Add verified seed 43 and 44 checkpoints

Adds the recovered best and final weights for seeds 43 and 44, with configs, histories, metrics, loading notes, and updated checksums.

README.md

@@ -122,4 +122,4 @@ released under the MIT License; these artifacts are derived from the surveys abo
 
 A dated release of the paper experiments is available under [`paper_experiments/2026-09-13/`](paper_experiments/2026-09-13/README.md). It includes the full-test bidirectional retrieval results, three-seed summary, narrow-redshift retrieval with paired bootstrap intervals, resource telemetry, spatial-split audit, proposed cell-disjoint manifest, analysis scripts, and checksum-verified evidence archives.
 
-The published seed-42 checkpoint remains the primary checkpoint in `full_run_110k/checkpoints/best/`. Metrics and provenance for the new seed 43 and 44 runs are included in the release. Their model weights are listed as pending until the detached RunPod volume can be attached and the uploaded files can be verified against the recorded hashes.
+The published seed-42 checkpoint remains the primary checkpoint in `full_run_110k/checkpoints/best/`. The release includes the new seed 43 and 44 best and final checkpoints, exact configs, histories, metrics, and verification hashes.

paper_experiments/2026-09-13/CHECKSUMS.sha256

@@ -1,10 +1,25 @@
+087e71cd8669584e60d77f810ccb43c48a966aa3224e5daf2ef29d6d03aa9b9a  checkpoints/README.md
+3d5d8be23cc7a129263235f299bcc39f3c6de003bd0c16734727707bb698dd42  checkpoints/seed43/best/meta.json
+5951b5617447563ea514ebce215b946015f7e9d1315eb5c26db0712cec2077a6  checkpoints/seed43/best/model.safetensors
+2cf90d29e699b93f66e5aa03ecd72ae9c0637c71a044fec7f0713a8ecc837c1c  checkpoints/seed43/config.yaml
+ddca16b7b65b3a0ba9987a49c510c2bb92399e9c0ed241b4f7f09b5abccfb6ff  checkpoints/seed43/evaluation.json
+b841cb60f5c0213fc0bfa342b587b296b212b7a60c65f690207df8ba184a71c2  checkpoints/seed43/final/meta.json
+07a6ced8c72535cd20240f0c7d04f09e131ef9447d6b0604b370e515f337073e  checkpoints/seed43/final/model.safetensors
+5ebffe1013f126a406ca37013aadc73f4ad74f959c0b73e43b65870732387c47  checkpoints/seed43/history.jsonl
+6bc5958599636442a8d9e903e3731bd58ecdf18629be1e786747bfabc76d228f  checkpoints/seed44/best/meta.json
+d5b20bc048de14602a91a3ee8730ef0b059a2a091083b5bd631fef9a95b39fdf  checkpoints/seed44/best/model.safetensors
+6f07d4e7de4c1adef0d7abb24e8422c08fa86c14a66e0d662e630aedb04de136  checkpoints/seed44/config.yaml
+4913a64ab337686005017f3248d3dfe88fc1ae8cc78ee89d7e688846ab473ee0  checkpoints/seed44/evaluation.json
+9266a4e4afd51aacb95f87d4120acd31e3d7deb334c36fa686395a4ca3299d9b  checkpoints/seed44/final/meta.json
+f81347a7cfb6694850d461140d58cd586d94d66e585c2836b08b9b9fb4009f66  checkpoints/seed44/final/model.safetensors
+cdd4c32427d8c6d335c5b308cb06701faf938400cb96f1462387fa4b94b655ad  checkpoints/seed44/history.jsonl
 4400b234702b60e36bf690414a42882c383faac96f83fdf6e509c58ede415118  evidence/cpu-diagnostics-evidence.tar.gz
 0abfa718a0e617a197505647cb04e0b36f2db98fb713d90fe40bdba1bb50d89e  evidence/gpu-results-evidence.tar.gz
 7ba109b7c000f6b2af62f1d4231ade72664b1bc5e475a5a54eea2990d864e95f  protocols/close_neighbor_redshift_audit.json
 73a5669e989f995b76ddcb25c5d3cb02e0a1fbd78912aac64ac539556213c11a  protocols/spatial_manifest_nside64.jsonl
 ef394dc8c03314c27840527999bf21db60628f49f88533d897b6cd6ee1eeed28  protocols/spatial_split_proposal.json
-a02b3ce547f0865f8ef16c7a7daa112127931037db39707722e29022dd4aa643  README.md
-ac846552a76ae80dc3c5ce512b29206c942221c0a333ff884b2e5f2491382baa  release.json
+7fba9afc3795af00976544688487e7c5e8375424330ed9f44a46cb124d5edcd3  README.md
+25097d71e6fae7d43f9d026b8b32b320324c8981c461308a0e145ddaf851fe95  release.json
 807b79df89fe58e8ca5dfc3ca0d6cfadcaef8d9b3c36b3e852b0fef30f0b8957  reports/cpu_robustness_diagnostics.md
 ede562274faa4248bcef88a5669120cc30bdac49c4d88c6836f2931dd3e00604  reports/gpu_results.md
 bb2d5bf09a0e6e660e3e4cf39572eab81ac022fe9c9968a12ef170a54241894d  scripts/aggregate_gpu_resources.py

paper_experiments/2026-09-13/README.md

@@ -4,6 +4,7 @@ This directory records the resource-bounded experiments run for the paper based
 
 ## Contents
 
+- `checkpoints/` contains the seed 43 and 44 best and final weights, exact configs, histories, metrics, and loading notes.
 - `reports/` contains readable accounts of the GPU experiment and the subsequent CPU robustness checks.
 - `tables/` contains run-level results, aggregate statistics, resource measurements, and the experiment registry.
 - `protocols/` contains the proposed cell-disjoint spatial manifest and the audit of close cross-split neighbours.
@@ -25,15 +26,10 @@ The original IID manifest contains no repeated object IDs across splits. It does
 
 The published seed-42 checkpoint remains under `full_run_110k/checkpoints/best/` at the repository root.
 
-The seed 43 and seed 44 model weights are **not included in this initial release**. Their metrics, histories, configurations, checkpoint metadata, and SHA-256 hashes are preserved in the evidence archive, but the weights remained on the detached RunPod volume when this release was prepared. They will be added after that volume is attached to a temporary pod and the uploaded hashes are verified.
-
-Expected best-checkpoint hashes:
-
-- seed 43: `5951b5617447563ea514ebce215b946015f7e9d1315eb5c26db0712cec2077a6`
-- seed 44: `d5b20bc048de14602a91a3ee8730ef0b059a2a091083b5bd631fef9a95b39fdf`
+The seed 43 and seed 44 best and final checkpoints are included under `checkpoints/`, together with their exact configs, validation histories, evaluation metrics, and metadata. All four weight files were recovered from the RunPod persistent volume and verified against the hashes recorded at training time. See [`checkpoints/README.md`](checkpoints/README.md) for loading instructions and the complete hash table.
 
 ## Interpretation
 
 The current evidence supports lightweight cross-modal alignment under a modest compute budget. It does not yet establish spatial generalisation or superiority to an astronomy-specific foundation model. The next model runs should use the cell-disjoint split and include a same-split AstroCLIP comparison. Match-radius sensitivity and shifted-coordinate contamination also remain open because the cached feature set does not retain rejected cross-match candidates.
 
-Small numerical differences can occur across CPU and CUDA backends for nearly tied similarity scores. Tables in `reports/` identify the evaluation environment and should be cited with the candidate-pool size.
+Small numerical differences can occur across CPU and CUDA backends for nearly tied similarity scores. Tables in `reports/` identify the evaluation environment and should be cited with the candidate-pool size.

paper_experiments/2026-09-13/checkpoints/README.md

@@ -0,0 +1,47 @@
+# Seed 43 and 44 checkpoints
+
+These checkpoints complete the three-seed baseline reported in this release. Each checkpoint stores the trainable image projection, spectrum encoder, spectrum projection, and temperature. The image input is a cached 1,024-dimensional CLIP feature, as specified by `image_encoder.type: identity` in each configuration.
+
+## Directory structure
+
+Each seed directory contains:
+
+- `config.yaml`: the exact training and evaluation configuration;
+- `best/`: the checkpoint selected by validation loss;
+- `final/`: the checkpoint at step 12,900;
+- `history.jsonl`: periodic training and validation records;
+- `evaluation.json`: full-test bidirectional retrieval and redshift-probe results for the best checkpoint.
+
+## Loading a checkpoint
+
+Use the code at commit `bda764c273e6663350a1d81a08bdedde119b0714`:
+
+```python
+from common import build_model, load_config
+from training.checkpoint import load_checkpoint
+import torch
+
+config = load_config("paper_experiments/2026-09-13/checkpoints/seed43/config.yaml")
+config["model"]["device"] = "cpu"  # change to "cuda" when available
+device = torch.device(config["model"]["device"])
+model = build_model(config, device)
+load_checkpoint(
+    model,
+    "paper_experiments/2026-09-13/checkpoints/seed43/best",
+    map_location=str(device),
+)
+model.eval()
+```
+
+The supplied configs retain the RunPod data and output paths used for provenance. Set `data.aligned_dir`, `training.output_dir`, and `eval.metrics_dir` to local paths before rerunning training or evaluation.
+
+## Recorded hashes
+
+| Seed | Checkpoint | Step | SHA-256 |
+|---|---|---:|---|
+| 43 | best | 5,000 | `5951b5617447563ea514ebce215b946015f7e9d1315eb5c26db0712cec2077a6` |
+| 43 | final | 12,900 | `07a6ced8c72535cd20240f0c7d04f09e131ef9447d6b0604b370e515f337073e` |
+| 44 | best | 4,000 | `d5b20bc048de14602a91a3ee8730ef0b059a2a091083b5bd631fef9a95b39fdf` |
+| 44 | final | 12,900 | `f81347a7cfb6694850d461140d58cd586d94d66e585c2836b08b9b9fb4009f66` |
+
+The release-level `CHECKSUMS.sha256` file covers these weights and all associated metadata.

paper_experiments/2026-09-13/checkpoints/seed43/best/meta.json

@@ -0,0 +1,5 @@
+{
+  "step": 5000,
+  "val_loss": 3.9103331565856934,
+  "val_recall@1": 0.02
+}

paper_experiments/2026-09-13/checkpoints/seed43/best/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5951b5617447563ea514ebce215b946015f7e9d1315eb5c26db0712cec2077a6
+size 6622652

paper_experiments/2026-09-13/checkpoints/seed43/config.yaml

@@ -0,0 +1,59 @@
+data:
+  aligned_dir: /workspace/artifacts/galaxy-image-spectrum-clip-ee2e189/precomputed_features
+  shard_cache_size: 0
+  image:
+    bands:
+    - g
+    - r
+    - z
+    image_size: 224
+    arcsinh_scale: 0.1
+  spectrum:
+    wavelength_min: 3600.0
+    wavelength_max: 9800.0
+    n_bins: 1024
+image_encoder:
+  type: identity
+  output_dim: 1024
+spectrum_encoder:
+  hidden_channels:
+  - 64
+  - 128
+  - 256
+  kernel_size: 5
+  embedding_dim: 512
+  dropout: 0.0
+  augment_noise_std: 0.0
+  augment_mask_frac: 0.0
+projection:
+  shared_dim: 512
+  hidden_dim: 512
+  dropout: 0.0
+model:
+  temperature_init: 0.07
+  device: cuda
+  torch_dtype: bfloat16
+training:
+  output_dir: /workspace/runs/gpu/baseline-seed43/checkpoints
+  num_epochs: 30
+  per_device_batch_size: 256
+  gradient_accumulation_steps: 1
+  learning_rate: 0.0005
+  warmup_ratio: 0.03
+  weight_decay: 0.05
+  max_grad_norm: 1.0
+  bf16: true
+  logging_steps: 20
+  eval_steps: 1000
+  save_steps: 0
+  eval_num_samples: 2000
+  dataloader_num_workers: 0
+  seed: 43
+eval:
+  recall_ks:
+  - 1
+  - 5
+  - 10
+  probe_targets:
+  - redshift
+  metrics_dir: /workspace/runs/gpu/baseline-seed43/metrics

paper_experiments/2026-09-13/checkpoints/seed43/evaluation.json

@@ -0,0 +1,34 @@
+{
+  "retrieval": {
+    "n": 13762,
+    "recall@1": 0.003197209707891295,
+    "recall@5": 0.013006830402557767,
+    "recall@10": 0.023470425810202006,
+    "median_rank": 658.0,
+    "mrr": 0.013072696474946122,
+    "image_to_spectrum_recall@1": 0.003197209707891295,
+    "image_to_spectrum_recall@5": 0.013006830402557767,
+    "image_to_spectrum_recall@10": 0.023470425810202006,
+    "image_to_spectrum_median_rank": 658.0,
+    "image_to_spectrum_mrr": 0.013072696474946122,
+    "spectrum_to_image_recall@1": 0.0026885627089085888,
+    "spectrum_to_image_recall@5": 0.013079494259555298,
+    "spectrum_to_image_recall@10": 0.02274378724022671,
+    "spectrum_to_image_median_rank": 680.0,
+    "spectrum_to_image_mrr": 0.012551868337039146
+  },
+  "probe_redshift": {
+    "image": {
+      "r2": 0.6801643967628479,
+      "mse": 0.011084817349910736,
+      "n_train": 110098,
+      "n_test": 13762
+    },
+    "spectrum": {
+      "r2": 0.8224244117736816,
+      "mse": 0.006154388189315796,
+      "n_train": 110098,
+      "n_test": 13762
+    }
+  }
+}

paper_experiments/2026-09-13/checkpoints/seed43/final/meta.json

@@ -0,0 +1,6 @@
+{
+  "step": 12900,
+  "train_loss": 2.947625160217285,
+  "val_loss": 4.031911849975586,
+  "val_recall@1": 0.0165
+}

paper_experiments/2026-09-13/checkpoints/seed43/final/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07a6ced8c72535cd20240f0c7d04f09e131ef9447d6b0604b370e515f337073e
+size 6622652

paper_experiments/2026-09-13/checkpoints/seed43/history.jsonl

@@ -0,0 +1,658 @@
+{"event": "train", "step": 20, "loss": 5.5281822919845585, "learning_rate": 2.5839793281653746e-05, "elapsed_s": 9.409430980682373}
+{"event": "train", "step": 40, "loss": 5.351397776603699, "learning_rate": 5.167958656330749e-05, "elapsed_s": 9.890164136886597}
+{"event": "train", "step": 60, "loss": 5.10305233001709, "learning_rate": 7.751937984496124e-05, "elapsed_s": 10.38361120223999}
+{"event": "train", "step": 80, "loss": 4.945181131362915, "learning_rate": 0.00010335917312661498, "elapsed_s": 10.862939596176147}
+{"event": "train", "step": 100, "loss": 4.87975161075592, "learning_rate": 0.00012919896640826872, "elapsed_s": 11.353288888931274}
+{"event": "train", "step": 120, "loss": 4.839201831817627, "learning_rate": 0.0001550387596899225, "elapsed_s": 11.827314376831055}
+{"event": "train", "step": 140, "loss": 4.77836422920227, "learning_rate": 0.00018087855297157623, "elapsed_s": 12.298125982284546}
+{"event": "train", "step": 160, "loss": 4.78362340927124, "learning_rate": 0.00020671834625322997, "elapsed_s": 12.790607213973999}
+{"event": "train", "step": 180, "loss": 4.701811623573303, "learning_rate": 0.00023255813953488373, "elapsed_s": 13.260878324508667}
+{"event": "train", "step": 200, "loss": 4.666235756874085, "learning_rate": 0.00025839793281653745, "elapsed_s": 13.758657693862915}
+{"event": "train", "step": 220, "loss": 4.637979912757873, "learning_rate": 0.0002842377260981912, "elapsed_s": 14.241762161254883}
+{"event": "train", "step": 240, "loss": 4.609049963951111, "learning_rate": 0.000310077519379845, "elapsed_s": 14.720380306243896}
+{"event": "train", "step": 260, "loss": 4.626545381546021, "learning_rate": 0.0003359173126614987, "elapsed_s": 15.220694065093994}
+{"event": "train", "step": 280, "loss": 4.612429618835449, "learning_rate": 0.00036175710594315246, "elapsed_s": 15.692527770996094}
+{"event": "train", "step": 300, "loss": 4.620847845077515, "learning_rate": 0.00038759689922480625, "elapsed_s": 16.187878608703613}
+{"event": "train", "step": 320, "loss": 4.576876521110535, "learning_rate": 0.00041343669250645994, "elapsed_s": 16.66387629508972}
+{"event": "train", "step": 340, "loss": 4.5739531993865965, "learning_rate": 0.0004392764857881137, "elapsed_s": 17.143698692321777}
+{"event": "train", "step": 360, "loss": 4.58502426147461, "learning_rate": 0.00046511627906976747, "elapsed_s": 17.63378930091858}
+{"event": "train", "step": 380, "loss": 4.599628973007202, "learning_rate": 0.0004909560723514212, "elapsed_s": 18.103389978408813}
+{"event": "train", "step": 400, "loss": 4.595144844055175, "learning_rate": 0.000499998668401834, "elapsed_s": 18.58367371559143}
+{"event": "train", "step": 420, "loss": 4.558658647537231, "learning_rate": 0.0004999914195065388, "elapsed_s": 19.062133073806763}
+{"event": "train", "step": 440, "loss": 4.491467404365539, "learning_rate": 0.0004999778674119666, "elapsed_s": 19.54169487953186}
+{"event": "train", "step": 460, "loss": 4.478895092010498, "learning_rate": 0.0004999580124598156, "elapsed_s": 20.012390851974487}
+{"event": "train", "step": 480, "loss": 4.426129961013794, "learning_rate": 0.0004999318551507018, "elapsed_s": 20.47928023338318}
+{"event": "train", "step": 500, "loss": 4.448922181129456, "learning_rate": 0.0004998993961441468, "elapsed_s": 20.94507646560669}
+{"event": "train", "step": 520, "loss": 4.427977752685547, "learning_rate": 0.0004998606362585609, "elapsed_s": 21.55736517906189}
+{"event": "train", "step": 540, "loss": 4.4980087041854855, "learning_rate": 0.000499815576471223, "elapsed_s": 22.023704767227173}
+{"event": "train", "step": 560, "loss": 4.4172645807266235, "learning_rate": 0.0004997642179182553, "elapsed_s": 22.487655639648438}
+{"event": "train", "step": 580, "loss": 4.401033186912537, "learning_rate": 0.0004997065618945947, "elapsed_s": 22.951995134353638}
+{"event": "train", "step": 600, "loss": 4.419301557540893, "learning_rate": 0.0004996426098539612, "elapsed_s": 23.418088912963867}
+{"event": "train", "step": 620, "loss": 4.4061686277389525, "learning_rate": 0.0004995723634088196, "elapsed_s": 23.87713646888733}
+{"event": "train", "step": 640, "loss": 4.357074356079101, "learning_rate": 0.0004994958243303403, "elapsed_s": 24.344048500061035}
+{"event": "train", "step": 660, "loss": 4.388646173477173, "learning_rate": 0.0004994129945483537, "elapsed_s": 24.806560039520264}
+{"event": "train", "step": 680, "loss": 4.384529066085816, "learning_rate": 0.0004993238761513019, "elapsed_s": 25.266124725341797}
+{"event": "train", "step": 700, "loss": 4.357813763618469, "learning_rate": 0.0004992284713861862, "elapsed_s": 25.72800302505493}
+{"event": "train", "step": 720, "loss": 4.353272128105163, "learning_rate": 0.0004991267826585101, "elapsed_s": 26.18946647644043}
+{"event": "train", "step": 740, "loss": 4.373588633537293, "learning_rate": 0.0004990188125322189, "elapsed_s": 26.649288415908813}
+{"event": "train", "step": 760, "loss": 4.363992714881897, "learning_rate": 0.0004989045637296348, "elapsed_s": 27.106199264526367}
+{"event": "train", "step": 780, "loss": 4.364336681365967, "learning_rate": 0.0004987840391313886, "elapsed_s": 27.571012020111084}
+{"event": "train", "step": 800, "loss": 4.325901770591736, "learning_rate": 0.000498657241776347, "elapsed_s": 28.02659821510315}
+{"event": "train", "step": 820, "loss": 4.350520896911621, "learning_rate": 0.0004985241748615355, "elapsed_s": 28.49514150619507}
+{"event": "train", "step": 840, "loss": 4.3536540269851685, "learning_rate": 0.0004983848417420585, "elapsed_s": 28.965089321136475}
+{"event": "train", "step": 860, "loss": 4.358001732826233, "learning_rate": 0.0004982392459310141, "elapsed_s": 29.420042991638184}
+{"event": "train", "step": 880, "loss": 4.239088177680969, "learning_rate": 0.0004980873910994061, "elapsed_s": 29.900376796722412}
+{"event": "train", "step": 900, "loss": 4.279656648635864, "learning_rate": 0.000497929281076051, "elapsed_s": 30.358550548553467}
+{"event": "train", "step": 920, "loss": 4.2728898286819454, "learning_rate": 0.0004977649198474814, "elapsed_s": 30.822954654693604}
+{"event": "train", "step": 940, "loss": 4.291485691070557, "learning_rate": 0.0004975943115578461, "elapsed_s": 31.277812242507935}
+{"event": "train", "step": 960, "loss": 4.23212206363678, "learning_rate": 0.0004974174605088046, "elapsed_s": 31.733376026153564}
+{"event": "train", "step": 980, "loss": 4.276410818099976, "learning_rate": 0.0004972343711594198, "elapsed_s": 32.185946226119995}
+{"event": "train", "step": 1000, "loss": 4.25772876739502, "learning_rate": 0.0004970450481260446, "elapsed_s": 32.637258529663086}
+{"event": "validation", "step": 1000, "loss": 4.295291244983673, "recall_at_1": 0.008}
+{"event": "train", "step": 1020, "loss": 4.2291077136993405, "learning_rate": 0.0004968494961822062, "elapsed_s": 41.3586266040802}
+{"event": "train", "step": 1040, "loss": 4.25956358909607, "learning_rate": 0.0004966477202584854, "elapsed_s": 41.81532859802246}
+{"event": "train", "step": 1060, "loss": 4.207193756103516, "learning_rate": 0.0004964397254423919, "elapsed_s": 42.26698899269104}
+{"event": "train", "step": 1080, "loss": 4.2378315925598145, "learning_rate": 0.000496225516978237, "elapsed_s": 42.717743158340454}
+{"event": "train", "step": 1100, "loss": 4.239974737167358, "learning_rate": 0.0004960051002670007, "elapsed_s": 43.16855525970459}
+{"event": "train", "step": 1120, "loss": 4.253215861320496, "learning_rate": 0.0004957784808661955, "elapsed_s": 43.61804676055908}
+{"event": "train", "step": 1140, "loss": 4.257088899612427, "learning_rate": 0.0004955456644897265, "elapsed_s": 44.06835603713989}
+{"event": "train", "step": 1160, "loss": 4.234441876411438, "learning_rate": 0.0004953066570077471, "elapsed_s": 44.523125886917114}
+{"event": "train", "step": 1180, "loss": 4.241541242599487, "learning_rate": 0.0004950614644465114, "elapsed_s": 44.979780435562134}
+{"event": "train", "step": 1200, "loss": 4.221798014640808, "learning_rate": 0.0004948100929882219, "elapsed_s": 45.42941904067993}
+{"event": "train", "step": 1220, "loss": 4.194061040878296, "learning_rate": 0.0004945525489708736, "elapsed_s": 45.879154205322266}
+{"event": "train", "step": 1240, "loss": 4.2020035743713375, "learning_rate": 0.0004942888388880946, "elapsed_s": 46.333412885665894}
+{"event": "train", "step": 1260, "loss": 4.227820920944214, "learning_rate": 0.0004940189693889818, "elapsed_s": 46.788278579711914}
+{"event": "train", "step": 1280, "loss": 4.19993302822113, "learning_rate": 0.0004937429472779336, "elapsed_s": 47.242366790771484}
+{"event": "train", "step": 1300, "loss": 4.184213924407959, "learning_rate": 0.0004934607795144786, "elapsed_s": 47.70129609107971}
+{"event": "train", "step": 1320, "loss": 4.157126379013062, "learning_rate": 0.0004931724732130993, "elapsed_s": 48.14993762969971}
+{"event": "train", "step": 1340, "loss": 4.166290581226349, "learning_rate": 0.0004928780356430537, "elapsed_s": 48.597355127334595}
+{"event": "train", "step": 1360, "loss": 4.179386305809021, "learning_rate": 0.0004925774742281911, "elapsed_s": 49.04364228248596}
+{"event": "train", "step": 1380, "loss": 4.172819066047668, "learning_rate": 0.0004922707965467657, "elapsed_s": 49.49231839179993}
+{"event": "train", "step": 1400, "loss": 4.15705189704895, "learning_rate": 0.0004919580103312449, "elapsed_s": 49.94084429740906}
+{"event": "train", "step": 1420, "loss": 4.1657795906066895, "learning_rate": 0.000491639123468115, "elapsed_s": 50.39029574394226}
+{"event": "train", "step": 1440, "loss": 4.157159423828125, "learning_rate": 0.0004913141439976816, "elapsed_s": 50.840710401535034}
+{"event": "train", "step": 1460, "loss": 4.15277783870697, "learning_rate": 0.0004909830801138673, "elapsed_s": 51.293541431427}
+{"event": "train", "step": 1480, "loss": 4.151175022125244, "learning_rate": 0.0004906459401640053, "elapsed_s": 51.74384951591492}
+{"event": "train", "step": 1500, "loss": 4.1283017039299015, "learning_rate": 0.0004903027326486286, "elapsed_s": 52.19592356681824}
+{"event": "train", "step": 1520, "loss": 4.119686841964722, "learning_rate": 0.0004899534662212553, "elapsed_s": 52.644770860672}
+{"event": "train", "step": 1540, "loss": 4.137971901893616, "learning_rate": 0.0004895981496881715, "elapsed_s": 53.103548526763916}
+{"event": "train", "step": 1560, "loss": 4.146236562728882, "learning_rate": 0.0004892367920082082, "elapsed_s": 53.55504536628723}
+{"event": "train", "step": 1580, "loss": 4.190236663818359, "learning_rate": 0.0004888694022925161, "elapsed_s": 54.00462889671326}
+{"event": "train", "step": 1600, "loss": 4.14671676158905, "learning_rate": 0.0004884959898043354, "elapsed_s": 54.45481634140015}
+{"event": "train", "step": 1620, "loss": 4.114955043792724, "learning_rate": 0.0004881165639587627, "elapsed_s": 54.91207671165466}
+{"event": "train", "step": 1640, "loss": 4.134228014945984, "learning_rate": 0.00048773113432251295, "elapsed_s": 55.364521980285645}
+{"event": "train", "step": 1660, "loss": 4.1087362408638, "learning_rate": 0.0004873397106136791, "elapsed_s": 55.81471133232117}
+{"event": "train", "step": 1680, "loss": 4.1336332082748415, "learning_rate": 0.0004869423027014864, "elapsed_s": 56.26461696624756}
+{"event": "train", "step": 1700, "loss": 4.1286639928817745, "learning_rate": 0.0004865389206060438, "elapsed_s": 56.712717056274414}
+{"event": "train", "step": 1720, "loss": 4.123653924465179, "learning_rate": 0.00048612957449809137, "elapsed_s": 57.15244507789612}
+{"event": "train", "step": 1740, "loss": 4.058912622928619, "learning_rate": 0.00048571427469874356, "elapsed_s": 57.62162923812866}
+{"event": "train", "step": 1760, "loss": 4.0701998472213745, "learning_rate": 0.0004852930316792296, "elapsed_s": 58.06824016571045}
+{"event": "train", "step": 1780, "loss": 4.100683116912842, "learning_rate": 0.0004848658560606285, "elapsed_s": 58.5149359703064}
+{"event": "train", "step": 1800, "loss": 4.101222062110901, "learning_rate": 0.00048443275861360236, "elapsed_s": 58.9608268737793}
+{"event": "train", "step": 1820, "loss": 4.115538430213928, "learning_rate": 0.0004839937502581239, "elapsed_s": 59.42372989654541}
+{"event": "train", "step": 1840, "loss": 4.04584310054779, "learning_rate": 0.00048354884206320176, "elapsed_s": 59.91778349876404}
+{"event": "train", "step": 1860, "loss": 4.083363389968872, "learning_rate": 0.00048309804524660096, "elapsed_s": 60.38676428794861}
+{"event": "train", "step": 1880, "loss": 4.119002366065979, "learning_rate": 0.0004826413711745602, "elapsed_s": 60.84481692314148}
+{"event": "train", "step": 1900, "loss": 4.140379416942596, "learning_rate": 0.00048217883136150545, "elapsed_s": 61.30591320991516}
+{"event": "train", "step": 1920, "loss": 4.096714615821838, "learning_rate": 0.0004817104374697595, "elapsed_s": 61.76745295524597}
+{"event": "train", "step": 1940, "loss": 4.1147495865821835, "learning_rate": 0.00048123620130924777, "elapsed_s": 62.22875261306763}
+{"event": "train", "step": 1960, "loss": 4.080975127220154, "learning_rate": 0.0004807561348372008, "elapsed_s": 62.71122622489929}
+{"event": "train", "step": 1980, "loss": 4.029158663749695, "learning_rate": 0.0004802702501578526, "elapsed_s": 63.17241954803467}
+{"event": "train", "step": 2000, "loss": 4.062751221656799, "learning_rate": 0.0004797785595221354, "elapsed_s": 63.63932394981384}
+{"event": "validation", "step": 2000, "loss": 4.028339207172394, "recall_at_1": 0.0125}
+{"event": "train", "step": 2020, "loss": 4.04496408700943, "learning_rate": 0.0004792810753273711, "elapsed_s": 64.444167137146}
+{"event": "train", "step": 2040, "loss": 4.0703299880027775, "learning_rate": 0.0004787778101169581, "elapsed_s": 64.90849232673645}
+{"event": "train", "step": 2060, "loss": 4.129874587059021, "learning_rate": 0.00047826877658005564, "elapsed_s": 65.38292527198792}
+{"event": "train", "step": 2080, "loss": 4.059052729606629, "learning_rate": 0.0004777539875512635, "elapsed_s": 65.85321736335754}
+{"event": "train", "step": 2100, "loss": 4.070672059059143, "learning_rate": 0.00047723345601029856, "elapsed_s": 66.32293224334717}
+{"event": "train", "step": 2120, "loss": 4.057565820217133, "learning_rate": 0.0004767071950816673, "elapsed_s": 66.78368878364563}
+{"event": "train", "step": 2140, "loss": 4.060412085056305, "learning_rate": 0.0004761752180343353, "elapsed_s": 67.24256825447083}
+{"event": "train", "step": 2160, "loss": 4.054379558563232, "learning_rate": 0.0004756375382813921, "elapsed_s": 67.70782327651978}
+{"event": "train", "step": 2180, "loss": 4.00014306306839, "learning_rate": 0.0004750941693797137, "elapsed_s": 68.29656624794006}
+{"event": "train", "step": 2200, "loss": 4.042799174785614, "learning_rate": 0.0004745451250296201, "elapsed_s": 68.74956369400024}
+{"event": "train", "step": 2220, "loss": 4.015317010879516, "learning_rate": 0.0004739904190745303, "elapsed_s": 69.20397138595581}
+{"event": "train", "step": 2240, "loss": 3.996222162246704, "learning_rate": 0.00047343006550061296, "elapsed_s": 69.66201877593994}
+{"event": "train", "step": 2260, "loss": 4.042374169826507, "learning_rate": 0.000472864078436434, "elapsed_s": 70.12157964706421}
+{"event": "train", "step": 2280, "loss": 4.003713476657867, "learning_rate": 0.0004722924721526002, "elapsed_s": 70.5796468257904}
+{"event": "train", "step": 2300, "loss": 4.037427282333374, "learning_rate": 0.0004717152610613995, "elapsed_s": 71.04144835472107}
+{"event": "train", "step": 2320, "loss": 4.025668644905091, "learning_rate": 0.0004711324597164376, "elapsed_s": 71.49929594993591}
+{"event": "train", "step": 2340, "loss": 4.015722119808197, "learning_rate": 0.0004705440828122708, "elapsed_s": 71.95248603820801}
+{"event": "train", "step": 2360, "loss": 4.0154869794845585, "learning_rate": 0.00046995014518403576, "elapsed_s": 72.43445014953613}
+{"event": "train", "step": 2380, "loss": 4.032552313804627, "learning_rate": 0.00046935066180707556, "elapsed_s": 72.8932044506073}
+{"event": "train", "step": 2400, "loss": 4.026321053504944, "learning_rate": 0.0004687456477965616, "elapsed_s": 73.35179924964905}
+{"event": "train", "step": 2420, "loss": 4.020947420597077, "learning_rate": 0.000468135118407113, "elapsed_s": 73.81023025512695}
+{"event": "train", "step": 2440, "loss": 4.033311605453491, "learning_rate": 0.0004675190890324116, "elapsed_s": 74.26693749427795}
+{"event": "train", "step": 2460, "loss": 4.036572635173798, "learning_rate": 0.00046689757520481425, "elapsed_s": 74.72959733009338}
+{"event": "train", "step": 2480, "loss": 4.0411598682403564, "learning_rate": 0.00046627059259496086, "elapsed_s": 75.21459889411926}
+{"event": "train", "step": 2500, "loss": 4.005121636390686, "learning_rate": 0.00046563815701137926, "elapsed_s": 75.69390964508057}
+{"event": "train", "step": 2520, "loss": 4.041015148162842, "learning_rate": 0.0004650002844000868, "elapsed_s": 76.15999579429626}
+{"event": "train", "step": 2540, "loss": 4.039835298061371, "learning_rate": 0.00046435699084418835, "elapsed_s": 76.62389349937439}
+{"event": "train", "step": 2560, "loss": 4.012939143180847, "learning_rate": 0.00046370829256347037, "elapsed_s": 77.08781933784485}
+{"event": "train", "step": 2580, "loss": 4.026614630222321, "learning_rate": 0.0004630542059139923, "elapsed_s": 77.54569435119629}
+{"event": "train", "step": 2600, "loss": 3.949051856994629, "learning_rate": 0.0004623947473876744, "elapsed_s": 78.04095244407654}
+{"event": "train", "step": 2620, "loss": 3.9982136845588685, "learning_rate": 0.00046172993361188113, "elapsed_s": 78.52057456970215}
+{"event": "train", "step": 2640, "loss": 3.9981759786605835, "learning_rate": 0.00046105978134900285, "elapsed_s": 78.98649191856384}
+{"event": "train", "step": 2660, "loss": 3.9929186701774597, "learning_rate": 0.0004603843074960324, "elapsed_s": 79.44832730293274}
+{"event": "train", "step": 2680, "loss": 3.9360607981681826, "learning_rate": 0.0004597035290841396, "elapsed_s": 79.90817046165466}
+{"event": "train", "step": 2700, "loss": 3.951689624786377, "learning_rate": 0.00045901746327824155, "elapsed_s": 80.3844268321991}
+{"event": "train", "step": 2720, "loss": 3.963700366020203, "learning_rate": 0.00045832612737656985, "elapsed_s": 80.84825825691223}
+{"event": "train", "step": 2740, "loss": 3.9822076678276064, "learning_rate": 0.00045762953881023474, "elapsed_s": 81.31878709793091}
+{"event": "train", "step": 2760, "loss": 3.9685913920402527, "learning_rate": 0.0004569277151427851, "elapsed_s": 81.80831742286682}
+{"event": "train", "step": 2780, "loss": 3.9655125021934508, "learning_rate": 0.000456220674069766, "elapsed_s": 82.28020358085632}
+{"event": "train", "step": 2800, "loss": 3.9919437646865843, "learning_rate": 0.00045550843341827255, "elapsed_s": 82.7486002445221}
+{"event": "train", "step": 2820, "loss": 3.9725462913513185, "learning_rate": 0.0004547910111465, "elapsed_s": 83.21390008926392}
+{"event": "train", "step": 2840, "loss": 3.9931506276130677, "learning_rate": 0.0004540684253432915, "elapsed_s": 83.68015813827515}
+{"event": "train", "step": 2860, "loss": 4.014842057228089, "learning_rate": 0.00045334069422768147, "elapsed_s": 84.15134644508362}
+{"event": "train", "step": 2880, "loss": 3.9549938082695006, "learning_rate": 0.00045260783614843684, "elapsed_s": 84.62728977203369}
+{"event": "train", "step": 2900, "loss": 3.9915728092193605, "learning_rate": 0.0004518698695835936, "elapsed_s": 85.10067391395569}
+{"event": "train", "step": 2920, "loss": 3.972988283634186, "learning_rate": 0.0004511268131399919, "elapsed_s": 85.57712507247925}
+{"event": "train", "step": 2940, "loss": 3.9836616277694703, "learning_rate": 0.000450378685552806, "elapsed_s": 86.04319047927856}
+{"event": "train", "step": 2960, "loss": 3.9670968651771545, "learning_rate": 0.0004496255056850724, "elapsed_s": 86.5063247680664}
+{"event": "train", "step": 2980, "loss": 3.9662657499313356, "learning_rate": 0.00044886729252721413, "elapsed_s": 86.96944832801819}
+{"event": "train", "step": 3000, "loss": 3.967751717567444, "learning_rate": 0.00044810406519656175, "elapsed_s": 87.43322205543518}
+{"event": "validation", "step": 3000, "loss": 4.029622405767441, "recall_at_1": 0.0115}
+{"event": "train", "step": 3020, "loss": 3.919190287590027, "learning_rate": 0.0004473358429368717, "elapsed_s": 88.19628930091858}
+{"event": "train", "step": 3040, "loss": 3.9500224590301514, "learning_rate": 0.0004465626451178406, "elapsed_s": 88.65771102905273}
+{"event": "train", "step": 3060, "loss": 3.928187537193298, "learning_rate": 0.0004457844912346174, "elapsed_s": 89.11509227752686}
+{"event": "train", "step": 3080, "loss": 3.933859634399414, "learning_rate": 0.00044500140090731146, "elapsed_s": 89.57357716560364}
+{"event": "train", "step": 3100, "loss": 3.926680862903595, "learning_rate": 0.00044421339388049804, "elapsed_s": 90.0301296710968}
+{"event": "train", "step": 3120, "loss": 3.965007960796356, "learning_rate": 0.00044342049002272024, "elapsed_s": 90.48877358436584}
+{"event": "train", "step": 3140, "loss": 3.9257279515266417, "learning_rate": 0.00044262270932598835, "elapsed_s": 90.9465479850769}
+{"event": "train", "step": 3160, "loss": 3.926665890216827, "learning_rate": 0.00044182007190527573, "elapsed_s": 91.42184734344482}
+{"event": "train", "step": 3180, "loss": 3.920234489440918, "learning_rate": 0.00044101259799801126, "elapsed_s": 91.8864221572876}
+{"event": "train", "step": 3200, "loss": 3.93715044260025, "learning_rate": 0.0004402003079635695, "elapsed_s": 92.34833979606628}
+{"event": "train", "step": 3220, "loss": 3.9130229234695433, "learning_rate": 0.0004393832222827573, "elapsed_s": 92.81190490722656}
+{"event": "train", "step": 3240, "loss": 3.966219699382782, "learning_rate": 0.00043856136155729725, "elapsed_s": 93.2714912891388}
+{"event": "train", "step": 3260, "loss": 3.920989692211151, "learning_rate": 0.0004377347465093082, "elapsed_s": 93.73455214500427}
+{"event": "train", "step": 3280, "loss": 3.920439660549164, "learning_rate": 0.00043690339798078314, "elapsed_s": 94.19318079948425}
+{"event": "train", "step": 3300, "loss": 3.9488315343856812, "learning_rate": 0.0004360673369330631, "elapsed_s": 94.66060423851013}
+{"event": "train", "step": 3320, "loss": 3.911288094520569, "learning_rate": 0.00043522658444630926, "elapsed_s": 95.1268618106842}
+{"event": "train", "step": 3340, "loss": 3.9482327938079833, "learning_rate": 0.000434381161718971, "elapsed_s": 95.59097218513489}
+{"event": "train", "step": 3360, "loss": 3.929948353767395, "learning_rate": 0.00043353109006725175, "elapsed_s": 96.06042742729187}
+{"event": "train", "step": 3380, "loss": 3.9435068488121034, "learning_rate": 0.0004326763909245711, "elapsed_s": 96.524338722229}
+{"event": "train", "step": 3400, "loss": 3.9427104353904725, "learning_rate": 0.00043181708584102475, "elapsed_s": 96.98883843421936}
+{"event": "train", "step": 3420, "loss": 3.9317301511764526, "learning_rate": 0.0004309531964828412, "elapsed_s": 97.46111059188843}
+{"event": "train", "step": 3440, "loss": 3.9336114048957826, "learning_rate": 0.000430084744631835, "elapsed_s": 97.92008018493652}
+{"event": "train", "step": 3460, "loss": 3.869375991821289, "learning_rate": 0.00042921175218485816, "elapsed_s": 98.40481328964233}
+{"event": "train", "step": 3480, "loss": 3.8728907108306885, "learning_rate": 0.00042833424115324743, "elapsed_s": 98.8603355884552}
+{"event": "train", "step": 3500, "loss": 3.894847297668457, "learning_rate": 0.00042745223366226986, "elapsed_s": 99.31741738319397}
+{"event": "train", "step": 3520, "loss": 3.878554975986481, "learning_rate": 0.0004265657519505647, "elapsed_s": 99.77505421638489}
+{"event": "train", "step": 3540, "loss": 3.8815280556678773, "learning_rate": 0.0004256748183695826, "elapsed_s": 100.23445248603821}
+{"event": "train", "step": 3560, "loss": 3.9058278441429137, "learning_rate": 0.0004247794553830221, "elapsed_s": 100.69559001922607}
+{"event": "train", "step": 3580, "loss": 3.8867170453071593, "learning_rate": 0.0004238796855662634, "elapsed_s": 101.16444158554077}
+{"event": "train", "step": 3600, "loss": 3.901770520210266, "learning_rate": 0.00042297553160579906, "elapsed_s": 101.62680864334106}
+{"event": "train", "step": 3620, "loss": 3.920444357395172, "learning_rate": 0.00042206701629866173, "elapsed_s": 102.09002232551575}
+{"event": "train", "step": 3640, "loss": 3.885425364971161, "learning_rate": 0.0004211541625518499, "elapsed_s": 102.55147194862366}
+{"event": "train", "step": 3660, "loss": 3.8897652864456176, "learning_rate": 0.0004202369933817497, "elapsed_s": 103.01977872848511}
+{"event": "train", "step": 3680, "loss": 3.903170585632324, "learning_rate": 0.0004193155319135551, "elapsed_s": 103.4778344631195}
+{"event": "train", "step": 3700, "loss": 3.916122114658356, "learning_rate": 0.00041838980138068444, "elapsed_s": 103.9423017501831}
+{"event": "train", "step": 3720, "loss": 3.867608106136322, "learning_rate": 0.0004174598251241951, "elapsed_s": 104.4075665473938}
+{"event": "train", "step": 3740, "loss": 3.894065594673157, "learning_rate": 0.00041652562659219447, "elapsed_s": 104.86452889442444}
+{"event": "train", "step": 3760, "loss": 3.8667284727096556, "learning_rate": 0.00041558722933924896, "elapsed_s": 105.32183957099915}
+{"event": "train", "step": 3780, "loss": 3.881791925430298, "learning_rate": 0.00041464465702579036, "elapsed_s": 105.78912138938904}
+{"event": "train", "step": 3800, "loss": 3.8485711812973022, "learning_rate": 0.00041369793341751866, "elapsed_s": 106.24772477149963}
+{"event": "train", "step": 3820, "loss": 3.875931668281555, "learning_rate": 0.0004127470823848035, "elapsed_s": 106.70866775512695}
+{"event": "train", "step": 3840, "loss": 3.8982669830322267, "learning_rate": 0.0004117921279020819, "elapsed_s": 107.17478775978088}
+{"event": "train", "step": 3860, "loss": 3.8815783619880677, "learning_rate": 0.00041083309404725386, "elapsed_s": 107.63678526878357}
+{"event": "train", "step": 3880, "loss": 3.856695818901062, "learning_rate": 0.0004098700050010753, "elapsed_s": 108.21173310279846}
+{"event": "train", "step": 3900, "loss": 3.836848831176758, "learning_rate": 0.00040890288504654836, "elapsed_s": 108.67025375366211}
+{"event": "train", "step": 3920, "loss": 3.845408093929291, "learning_rate": 0.00040793175856830924, "elapsed_s": 109.12752270698547}
+{"event": "train", "step": 3940, "loss": 3.8372256755828857, "learning_rate": 0.0004069566500520131, "elapsed_s": 109.59024930000305}
+{"event": "train", "step": 3960, "loss": 3.821768116950989, "learning_rate": 0.00040597758408371707, "elapsed_s": 110.04481482505798}
+{"event": "train", "step": 3980, "loss": 3.828645038604736, "learning_rate": 0.0004049945853492599, "elapsed_s": 110.50065684318542}
+{"event": "train", "step": 4000, "loss": 3.8560713052749636, "learning_rate": 0.00040400767863363996, "elapsed_s": 110.95890021324158}
+{"event": "validation", "step": 4000, "loss": 3.952131360769272, "recall_at_1": 0.018}
+{"event": "train", "step": 4020, "loss": 3.840569865703583, "learning_rate": 0.00040301688882039014, "elapsed_s": 111.84181523323059}
+{"event": "train", "step": 4040, "loss": 3.8420878648757935, "learning_rate": 0.0004020222408909503, "elapsed_s": 112.33437824249268}
+{"event": "train", "step": 4060, "loss": 3.856982100009918, "learning_rate": 0.00040102375992403773, "elapsed_s": 112.79777312278748}
+{"event": "train", "step": 4080, "loss": 3.8411465883255005, "learning_rate": 0.0004000214710950145, "elapsed_s": 113.26255226135254}
+{"event": "train", "step": 4100, "loss": 3.844648778438568, "learning_rate": 0.0003990153996752528, "elapsed_s": 113.72133588790894}
+{"event": "train", "step": 4120, "loss": 3.843426060676575, "learning_rate": 0.0003980055710314979, "elapsed_s": 114.17719960212708}
+{"event": "train", "step": 4140, "loss": 3.8551520109176636, "learning_rate": 0.00039699201062522844, "elapsed_s": 114.63316607475281}
+{"event": "train", "step": 4160, "loss": 3.825196313858032, "learning_rate": 0.00039597474401201427, "elapsed_s": 115.08829736709595}
+{"event": "train", "step": 4180, "loss": 3.810824179649353, "learning_rate": 0.0003949537968408726, "elapsed_s": 115.54763841629028}
+{"event": "train", "step": 4200, "loss": 3.825093412399292, "learning_rate": 0.00039392919485362076, "elapsed_s": 116.01063656806946}
+{"event": "train", "step": 4220, "loss": 3.8148209691047668, "learning_rate": 0.00039290096388422743, "elapsed_s": 116.48033952713013}
+{"event": "train", "step": 4240, "loss": 3.873659098148346, "learning_rate": 0.00039186912985816126, "elapsed_s": 116.95984816551208}
+{"event": "train", "step": 4260, "loss": 3.8622162103652955, "learning_rate": 0.000390833718791737, "elapsed_s": 117.43173170089722}
+{"event": "train", "step": 4280, "loss": 3.883093774318695, "learning_rate": 0.00038979475679146007, "elapsed_s": 117.89131236076355}
+{"event": "train", "step": 4300, "loss": 3.8757270216941833, "learning_rate": 0.0003887522700533675, "elapsed_s": 118.33669686317444}
+{"event": "train", "step": 4320, "loss": 3.791212058067322, "learning_rate": 0.0003877062848623681, "elapsed_s": 118.82550692558289}
+{"event": "train", "step": 4340, "loss": 3.7624561309814455, "learning_rate": 0.00038665682759157945, "elapsed_s": 119.28023386001587}
+{"event": "train", "step": 4360, "loss": 3.7806105494499205, "learning_rate": 0.00038560392470166294, "elapsed_s": 119.73430490493774}
+{"event": "train", "step": 4380, "loss": 3.7972315192222594, "learning_rate": 0.0003845476027401568, "elapsed_s": 120.190349817276}
+{"event": "train", "step": 4400, "loss": 3.804866874217987, "learning_rate": 0.00038348788834080614, "elapsed_s": 120.64414644241333}
+{"event": "train", "step": 4420, "loss": 3.8174864768981935, "learning_rate": 0.0003824248082228925, "elapsed_s": 121.1018922328949}
+{"event": "train", "step": 4440, "loss": 3.791877341270447, "learning_rate": 0.000381358389190559, "elapsed_s": 121.55983638763428}
+{"event": "train", "step": 4460, "loss": 3.8366316199302672, "learning_rate": 0.0003802886581321353, "elapsed_s": 122.027019739151}
+{"event": "train", "step": 4480, "loss": 3.8020743250846865, "learning_rate": 0.0003792156420194591, "elapsed_s": 122.49603915214539}
+{"event": "train", "step": 4500, "loss": 3.756265377998352, "learning_rate": 0.0003781393679071967, "elapsed_s": 122.96347689628601}
+{"event": "train", "step": 4520, "loss": 3.802548956871033, "learning_rate": 0.0003770598629321602, "elapsed_s": 123.43565797805786}
+{"event": "train", "step": 4540, "loss": 3.7748663663864135, "learning_rate": 0.0003759771543126237, "elapsed_s": 123.89718174934387}
+{"event": "train", "step": 4560, "loss": 3.809177613258362, "learning_rate": 0.00037489126934763694, "elapsed_s": 124.35996460914612}
+{"event": "train", "step": 4580, "loss": 3.825628900527954, "learning_rate": 0.0003738022354163369, "elapsed_s": 124.82773423194885}
+{"event": "train", "step": 4600, "loss": 3.8121306180953978, "learning_rate": 0.00037271007997725756, "elapsed_s": 125.3428246974945}
+{"event": "train", "step": 4620, "loss": 3.801201045513153, "learning_rate": 0.0003716148305676376, "elapsed_s": 125.86530661582947}
+{"event": "train", "step": 4640, "loss": 3.7947699189186097, "learning_rate": 0.00037051651480272594, "elapsed_s": 126.37824535369873}
+{"event": "train", "step": 4660, "loss": 3.7910957455635073, "learning_rate": 0.0003694151603750856, "elapsed_s": 126.84745597839355}
+{"event": "train", "step": 4680, "loss": 3.806348276138306, "learning_rate": 0.0003683107950538954, "elapsed_s": 127.30728244781494}
+{"event": "train", "step": 4700, "loss": 3.78053457736969, "learning_rate": 0.0003672034466842498, "elapsed_s": 127.76462769508362}
+{"event": "train", "step": 4720, "loss": 3.8258302450180053, "learning_rate": 0.0003660931431864567, "elapsed_s": 128.22376894950867}
+{"event": "train", "step": 4740, "loss": 3.7969159960746763, "learning_rate": 0.000364979912555334, "elapsed_s": 128.69571185112}
+{"event": "train", "step": 4760, "loss": 3.7307664155960083, "learning_rate": 0.0003638637828595031, "elapsed_s": 129.15345215797424}
+{"event": "train", "step": 4780, "loss": 3.7572096228599547, "learning_rate": 0.0003627447822406813, "elapsed_s": 129.61424136161804}
+{"event": "train", "step": 4800, "loss": 3.7580011248588563, "learning_rate": 0.00036162293891297257, "elapsed_s": 130.07202315330505}
+{"event": "train", "step": 4820, "loss": 3.7375194311141966, "learning_rate": 0.000360498281162156, "elapsed_s": 130.5288257598877}
+{"event": "train", "step": 4840, "loss": 3.730088496208191, "learning_rate": 0.00035937083734497246, "elapsed_s": 130.9844512939453}
+{"event": "train", "step": 4860, "loss": 3.7572239875793456, "learning_rate": 0.0003582406358884098, "elapsed_s": 131.43931436538696}
+{"event": "train", "step": 4880, "loss": 3.764478850364685, "learning_rate": 0.0003571077052889863, "elapsed_s": 131.8996353149414}
+{"event": "train", "step": 4900, "loss": 3.7746726632118226, "learning_rate": 0.00035597207411203153, "elapsed_s": 132.3605670928955}
+{"event": "train", "step": 4920, "loss": 3.74833801984787, "learning_rate": 0.0003548337709909667, "elapsed_s": 132.8163447380066}
+{"event": "train", "step": 4940, "loss": 3.74993292093277, "learning_rate": 0.0003536928246265826, "elapsed_s": 133.26911902427673}
+{"event": "train", "step": 4960, "loss": 3.76016845703125, "learning_rate": 0.00035254926378631567, "elapsed_s": 133.72342824935913}
+{"event": "train", "step": 4980, "loss": 3.735177981853485, "learning_rate": 0.00035140311730352315, "elapsed_s": 134.17867302894592}
+{"event": "train", "step": 5000, "loss": 3.75859797000885, "learning_rate": 0.0003502544140767555, "elapsed_s": 134.63353323936462}
+{"event": "validation", "step": 5000, "loss": 3.9103331565856934, "recall_at_1": 0.02}
+{"event": "train", "step": 5020, "loss": 3.7426064014434814, "learning_rate": 0.0003491031830690282, "elapsed_s": 135.43403244018555}
+{"event": "train", "step": 5040, "loss": 3.7711474180221556, "learning_rate": 0.00034794945330709126, "elapsed_s": 135.891273021698}
+{"event": "train", "step": 5060, "loss": 3.7517244696617125, "learning_rate": 0.0003467932538806976, "elapsed_s": 136.350900888443}
+{"event": "train", "step": 5080, "loss": 3.7690706133842466, "learning_rate": 0.00034563461394186934, "elapsed_s": 136.80932521820068}
+{"event": "train", "step": 5100, "loss": 3.772673177719116, "learning_rate": 0.0003444735627041627, "elapsed_s": 137.26786041259766}
+{"event": "train", "step": 5120, "loss": 3.7662983298301698, "learning_rate": 0.0003433101294419319, "elapsed_s": 137.72632312774658}
+{"event": "train", "step": 5140, "loss": 3.744901168346405, "learning_rate": 0.00034214434348959046, "elapsed_s": 138.18526458740234}
+{"event": "train", "step": 5160, "loss": 3.774168860912323, "learning_rate": 0.0003409762342408719, "elapsed_s": 138.6369812488556}
+{"event": "train", "step": 5180, "loss": 3.7008399963378906, "learning_rate": 0.00033980583114808855, "elapsed_s": 139.12149906158447}
+{"event": "train", "step": 5200, "loss": 3.6592372179031374, "learning_rate": 0.00033863316372138906, "elapsed_s": 139.57922887802124}
+{"event": "train", "step": 5220, "loss": 3.712941515445709, "learning_rate": 0.0003374582615280141, "elapsed_s": 140.03467869758606}
+{"event": "train", "step": 5240, "loss": 3.7048124074935913, "learning_rate": 0.000336281154191551, "elapsed_s": 140.49220204353333}
+{"event": "train", "step": 5260, "loss": 3.7182647705078127, "learning_rate": 0.0003351018713911869, "elapsed_s": 140.9475495815277}
+{"event": "train", "step": 5280, "loss": 3.7008349537849425, "learning_rate": 0.0003339204428609602, "elapsed_s": 141.41062760353088}
+{"event": "train", "step": 5300, "loss": 3.715847301483154, "learning_rate": 0.00033273689838901133, "elapsed_s": 141.88203048706055}
+{"event": "train", "step": 5320, "loss": 3.731283497810364, "learning_rate": 0.0003315512678168309, "elapsed_s": 142.33928108215332}
+{"event": "train", "step": 5340, "loss": 3.683084774017334, "learning_rate": 0.0003303635810385082, "elapsed_s": 142.80047917366028}
+{"event": "train", "step": 5360, "loss": 3.708747673034668, "learning_rate": 0.00032917386799997673, "elapsed_s": 143.2666597366333}
+{"event": "train", "step": 5380, "loss": 3.7267611742019655, "learning_rate": 0.0003279821586982594, "elapsed_s": 143.73017191886902}
+{"event": "train", "step": 5400, "loss": 3.725152063369751, "learning_rate": 0.00032678848318071246, "elapsed_s": 144.19576263427734}
+{"event": "train", "step": 5420, "loss": 3.6886518239974975, "learning_rate": 0.0003255928715442674, "elapsed_s": 144.66077542304993}
+{"event": "train", "step": 5440, "loss": 3.7261673331260683, "learning_rate": 0.0003243953539346728, "elapsed_s": 145.1279685497284}
+{"event": "train", "step": 5460, "loss": 3.67485648393631, "learning_rate": 0.0003231959605457331, "elapsed_s": 145.59341192245483}
+{"event": "train", "step": 5480, "loss": 3.699032759666443, "learning_rate": 0.00032199472161854864, "elapsed_s": 146.06160831451416}
+{"event": "train", "step": 5500, "loss": 3.7231250286102293, "learning_rate": 0.00032079166744075215, "elapsed_s": 146.52705001831055}
+{"event": "train", "step": 5520, "loss": 3.6909382700920106, "learning_rate": 0.0003195868283457458, "elapsed_s": 146.9968056678772}
+{"event": "train", "step": 5540, "loss": 3.6932717323303224, "learning_rate": 0.0003183802347119358, "elapsed_s": 147.46106028556824}
+{"event": "train", "step": 5560, "loss": 3.7026296854019165, "learning_rate": 0.0003171719169619668, "elapsed_s": 147.92346119880676}
+{"event": "train", "step": 5580, "loss": 3.7627456903457643, "learning_rate": 0.0003159619055619547, "elapsed_s": 148.51610350608826}
+{"event": "train", "step": 5600, "loss": 3.676648771762848, "learning_rate": 0.0003147502310207186, "elapsed_s": 148.98367428779602}
+{"event": "train", "step": 5620, "loss": 3.636887526512146, "learning_rate": 0.0003135369238890114, "elapsed_s": 149.43767857551575}
+{"event": "train", "step": 5640, "loss": 3.6826322436332704, "learning_rate": 0.00031232201475874953, "elapsed_s": 149.89349842071533}
+{"event": "train", "step": 5660, "loss": 3.6390965461730955, "learning_rate": 0.0003111055342622419, "elapsed_s": 150.3481731414795}
+{"event": "train", "step": 5680, "loss": 3.6278772592544555, "learning_rate": 0.0003098875130714172, "elapsed_s": 150.7999665737152}
+{"event": "train", "step": 5700, "loss": 3.611040937900543, "learning_rate": 0.0003086679818970504, "elapsed_s": 151.25445342063904}
+{"event": "train", "step": 5720, "loss": 3.626362407207489, "learning_rate": 0.0003074469714879892, "elapsed_s": 151.70971775054932}
+{"event": "train", "step": 5740, "loss": 3.687831175327301, "learning_rate": 0.00030622451263037783, "elapsed_s": 152.1669156551361}
+{"event": "train", "step": 5760, "loss": 3.649583876132965, "learning_rate": 0.00030500063614688136, "elapsed_s": 152.63748121261597}
+{"event": "train", "step": 5780, "loss": 3.612541711330414, "learning_rate": 0.0003037753728959082, "elapsed_s": 153.09702944755554}
+{"event": "train", "step": 5800, "loss": 3.681780242919922, "learning_rate": 0.0003025487537708327, "elapsed_s": 153.55623173713684}
+{"event": "train", "step": 5820, "loss": 3.646482837200165, "learning_rate": 0.0003013208096992154, "elapsed_s": 154.01156449317932}
+{"event": "train", "step": 5840, "loss": 3.6706498980522158, "learning_rate": 0.0003000915716420235, "elapsed_s": 154.47064471244812}
+{"event": "train", "step": 5860, "loss": 3.676071786880493, "learning_rate": 0.0002988610705928507, "elapsed_s": 154.93793272972107}
+{"event": "train", "step": 5880, "loss": 3.6769017577171326, "learning_rate": 0.0002976293375771351, "elapsed_s": 155.40616106987}
+{"event": "train", "step": 5900, "loss": 3.700552022457123, "learning_rate": 0.00029639640365137723, "elapsed_s": 155.87000393867493}
+{"event": "train", "step": 5920, "loss": 3.659304106235504, "learning_rate": 0.00029516229990235687, "elapsed_s": 156.3303668498993}
+{"event": "train", "step": 5940, "loss": 3.7185670495033265, "learning_rate": 0.0002939270574463495, "elapsed_s": 156.7903814315796}
+{"event": "train", "step": 5960, "loss": 3.708423948287964, "learning_rate": 0.00029269070742834156, "elapsed_s": 157.25959849357605}
+{"event": "train", "step": 5980, "loss": 3.665572941303253, "learning_rate": 0.0002914532810212451, "elapsed_s": 157.71580123901367}
+{"event": "train", "step": 6000, "loss": 3.6398338317871093, "learning_rate": 0.00029021480942511185, "elapsed_s": 158.18613696098328}
+{"event": "validation", "step": 6000, "loss": 3.9188173711299896, "recall_at_1": 0.014}
+{"event": "train", "step": 6020, "loss": 3.638670575618744, "learning_rate": 0.00028897532386634663, "elapsed_s": 158.8711588382721}
+{"event": "train", "step": 6040, "loss": 3.5870279908180236, "learning_rate": 0.00028773485559692, "elapsed_s": 159.35385274887085}
+{"event": "train", "step": 6060, "loss": 3.566163718700409, "learning_rate": 0.0002864934358935803, "elapsed_s": 159.8123471736908}
+{"event": "train", "step": 6080, "loss": 3.616424095630646, "learning_rate": 0.00028525109605706485, "elapsed_s": 160.269540309906}
+{"event": "train", "step": 6100, "loss": 3.57941073179245, "learning_rate": 0.00028400786741131104, "elapsed_s": 160.7274444103241}
+{"event": "train", "step": 6120, "loss": 3.593774402141571, "learning_rate": 0.0002827637813026663, "elapsed_s": 161.1845772266388}
+{"event": "train", "step": 6140, "loss": 3.5814071178436278, "learning_rate": 0.0002815188690990979, "elapsed_s": 161.64232659339905}
+{"event": "train", "step": 6160, "loss": 3.581125259399414, "learning_rate": 0.00028027316218940193, "elapsed_s": 162.10053372383118}
+{"event": "train", "step": 6180, "loss": 3.6090366721153258, "learning_rate": 0.0002790266919824121, "elapsed_s": 162.5717577934265}
+{"event": "train", "step": 6200, "loss": 3.609307277202606, "learning_rate": 0.00027777948990620737, "elapsed_s": 163.02686619758606}
+{"event": "train", "step": 6220, "loss": 3.5957985281944276, "learning_rate": 0.00027653158740732005, "elapsed_s": 163.4869568347931}
+{"event": "train", "step": 6240, "loss": 3.6178934812545775, "learning_rate": 0.0002752830159499424, "elapsed_s": 163.9432258605957}
+{"event": "train", "step": 6260, "loss": 3.5868804812431336, "learning_rate": 0.00027403380701513405, "elapsed_s": 164.41288375854492}
+{"event": "train", "step": 6280, "loss": 3.659030866622925, "learning_rate": 0.00027278399210002715, "elapsed_s": 164.87823390960693}
+{"event": "train", "step": 6300, "loss": 3.646504008769989, "learning_rate": 0.00027153360271703326, "elapsed_s": 165.33666253089905}
+{"event": "train", "step": 6320, "loss": 3.6121468782424926, "learning_rate": 0.0002702826703930483, "elapsed_s": 165.79165482521057}
+{"event": "train", "step": 6340, "loss": 3.611432933807373, "learning_rate": 0.0002690312266686577, "elapsed_s": 166.25528240203857}
+{"event": "train", "step": 6360, "loss": 3.599980914592743, "learning_rate": 0.0002677793030973411, "elapsed_s": 166.71124815940857}
+{"event": "train", "step": 6380, "loss": 3.675539231300354, "learning_rate": 0.0002665269312446768, "elapsed_s": 167.17999577522278}
+{"event": "train", "step": 6400, "loss": 3.628689444065094, "learning_rate": 0.00026527414268754615, "elapsed_s": 167.63486337661743}
+{"event": "train", "step": 6420, "loss": 3.6450837373733522, "learning_rate": 0.00026402096901333703, "elapsed_s": 168.09277415275574}
+{"event": "train", "step": 6440, "loss": 3.638055515289307, "learning_rate": 0.00026276744181914736, "elapsed_s": 168.54800581932068}
+{"event": "train", "step": 6460, "loss": 3.573882019519806, "learning_rate": 0.00026151359271098885, "elapsed_s": 169.03055715560913}
+{"event": "train", "step": 6480, "loss": 3.526366353034973, "learning_rate": 0.00026025945330298954, "elapsed_s": 169.48559713363647}
+{"event": "train", "step": 6500, "loss": 3.5328050374984743, "learning_rate": 0.0002590050552165975, "elapsed_s": 169.94286036491394}
+{"event": "train", "step": 6520, "loss": 3.539969778060913, "learning_rate": 0.00025775043007978245, "elapsed_s": 170.39717030525208}
+{"event": "train", "step": 6540, "loss": 3.5613261818885804, "learning_rate": 0.0002564956095262392, "elapsed_s": 170.85092091560364}
+{"event": "train", "step": 6560, "loss": 3.542152392864227, "learning_rate": 0.0002552406251945897, "elapsed_s": 171.30446887016296}
+{"event": "train", "step": 6580, "loss": 3.565830075740814, "learning_rate": 0.0002539855087275853, "elapsed_s": 171.76062774658203}
+{"event": "train", "step": 6600, "loss": 3.567234468460083, "learning_rate": 0.000252730291771309, "elapsed_s": 172.2238826751709}
+{"event": "train", "step": 6620, "loss": 3.515272045135498, "learning_rate": 0.0002514750059743775, "elapsed_s": 172.68764877319336}
+{"event": "train", "step": 6640, "loss": 3.5666486740112306, "learning_rate": 0.00025021968298714317, "elapsed_s": 173.1594009399414}
+{"event": "train", "step": 6660, "loss": 3.5582181453704833, "learning_rate": 0.0002489643544608962, "elapsed_s": 173.63165545463562}
+{"event": "train", "step": 6680, "loss": 3.579266810417175, "learning_rate": 0.0002477090520470662, "elapsed_s": 174.0925145149231}
+{"event": "train", "step": 6700, "loss": 3.58918936252594, "learning_rate": 0.0002464538073964247, "elapsed_s": 174.55287313461304}
+{"event": "train", "step": 6720, "loss": 3.5733452439308167, "learning_rate": 0.0002451986521582867, "elapsed_s": 175.01169991493225}
+{"event": "train", "step": 6740, "loss": 3.5472938299179075, "learning_rate": 0.00024394361797971261, "elapsed_s": 175.46656346321106}
+{"event": "train", "step": 6760, "loss": 3.562128412723541, "learning_rate": 0.00024268873650471073, "elapsed_s": 175.92373704910278}
+{"event": "train", "step": 6780, "loss": 3.546475279331207, "learning_rate": 0.000241434039373439, "elapsed_s": 176.38252639770508}
+{"event": "train", "step": 6800, "loss": 3.5401865243911743, "learning_rate": 0.00024017955822140742, "elapsed_s": 176.86021661758423}
+{"event": "train", "step": 6820, "loss": 3.5760895013809204, "learning_rate": 0.00023892532467868028, "elapsed_s": 177.31835770606995}
+{"event": "train", "step": 6840, "loss": 3.5524646401405335, "learning_rate": 0.00023767137036907892, "elapsed_s": 177.77722477912903}
+{"event": "train", "step": 6860, "loss": 3.599333393573761, "learning_rate": 0.00023641772690938405, "elapsed_s": 178.2375991344452}
+{"event": "train", "step": 6880, "loss": 3.582583487033844, "learning_rate": 0.0002351644259085387, "elapsed_s": 178.68139934539795}
+{"event": "train", "step": 6900, "loss": 3.5017905235290527, "learning_rate": 0.00023391149896685135, "elapsed_s": 179.1611053943634}
+{"event": "train", "step": 6920, "loss": 3.448912227153778, "learning_rate": 0.00023265897767519903, "elapsed_s": 179.6162133216858}
+{"event": "train", "step": 6940, "loss": 3.4967374444007873, "learning_rate": 0.00023140689361423084, "elapsed_s": 180.09576392173767}
+{"event": "train", "step": 6960, "loss": 3.504877030849457, "learning_rate": 0.00023015527835357167, "elapsed_s": 180.55318427085876}
+{"event": "train", "step": 6980, "loss": 3.477652382850647, "learning_rate": 0.00022890416345102633, "elapsed_s": 181.01161360740662}
+{"event": "train", "step": 7000, "loss": 3.49927214384079, "learning_rate": 0.00022765358045178367, "elapsed_s": 181.46936464309692}
+{"event": "validation", "step": 7000, "loss": 3.9235774278640747, "recall_at_1": 0.0125}
+{"event": "train", "step": 7020, "loss": 3.4923967838287355, "learning_rate": 0.0002264035608876213, "elapsed_s": 182.1902048587799}
+{"event": "train", "step": 7040, "loss": 3.463062584400177, "learning_rate": 0.00022515413627611076, "elapsed_s": 182.65381836891174}
+{"event": "train", "step": 7060, "loss": 3.5436992287635802, "learning_rate": 0.0002239053381198224, "elapsed_s": 183.14178705215454}
+{"event": "train", "step": 7080, "loss": 3.453770172595978, "learning_rate": 0.00022265719790553147, "elapsed_s": 183.60932183265686}
+{"event": "train", "step": 7100, "loss": 3.4967313766479493, "learning_rate": 0.00022140974710342401, "elapsed_s": 184.0744388103485}
+{"event": "train", "step": 7120, "loss": 3.53732146024704, "learning_rate": 0.00022016301716630367, "elapsed_s": 184.53376030921936}
+{"event": "train", "step": 7140, "loss": 3.5385849833488465, "learning_rate": 0.0002189170395287982, "elapsed_s": 184.9929804801941}
+{"event": "train", "step": 7160, "loss": 3.507464921474457, "learning_rate": 0.00021767184560656715, "elapsed_s": 185.45259976387024}
+{"event": "train", "step": 7180, "loss": 3.486584150791168, "learning_rate": 0.00021642746679551, "elapsed_s": 185.91098499298096}
+{"event": "train", "step": 7200, "loss": 3.482814848423004, "learning_rate": 0.00021518393447097402, "elapsed_s": 186.39101433753967}
+{"event": "train", "step": 7220, "loss": 3.505258059501648, "learning_rate": 0.00021394127998696348, "elapsed_s": 186.85408282279968}
+{"event": "train", "step": 7240, "loss": 3.525926649570465, "learning_rate": 0.00021269953467534906, "elapsed_s": 187.31587171554565}
+{"event": "train", "step": 7260, "loss": 3.5476800441741942, "learning_rate": 0.00021145872984507818, "elapsed_s": 187.7814712524414}
+{"event": "train", "step": 7280, "loss": 3.5200006484985353, "learning_rate": 0.0002102188967813849, "elapsed_s": 188.3575096130371}
+{"event": "train", "step": 7300, "loss": 3.491549003124237, "learning_rate": 0.00020898006674500156, "elapsed_s": 188.81574273109436}
+{"event": "train", "step": 7320, "loss": 3.488608717918396, "learning_rate": 0.00020774227097137074, "elapsed_s": 189.28341722488403}
+{"event": "train", "step": 7340, "loss": 3.439854371547699, "learning_rate": 0.0002065055406698573, "elapsed_s": 189.7421760559082}
+{"event": "train", "step": 7360, "loss": 3.4437950253486633, "learning_rate": 0.00020526990702296154, "elapsed_s": 190.1971516609192}
+{"event": "train", "step": 7380, "loss": 3.4218014597892763, "learning_rate": 0.00020403540118553315, "elapsed_s": 190.65039920806885}
+{"event": "train", "step": 7400, "loss": 3.375540089607239, "learning_rate": 0.00020280205428398584, "elapsed_s": 191.10628008842468}
+{"event": "train", "step": 7420, "loss": 3.4354086399078367, "learning_rate": 0.0002015698974155121, "elapsed_s": 191.56939578056335}
+{"event": "train", "step": 7440, "loss": 3.4152805328369142, "learning_rate": 0.00020033896164729908, "elapsed_s": 192.02570152282715}
+{"event": "train", "step": 7460, "loss": 3.4430277347564697, "learning_rate": 0.00019910927801574613, "elapsed_s": 192.49676370620728}
+{"event": "train", "step": 7480, "loss": 3.445837116241455, "learning_rate": 0.00019788087752568116, "elapsed_s": 192.9657862186432}
+{"event": "train", "step": 7500, "loss": 3.4863911747932432, "learning_rate": 0.00019665379114957955, "elapsed_s": 193.42709016799927}
+{"event": "train", "step": 7520, "loss": 3.447878730297089, "learning_rate": 0.00019542804982678285, "elapsed_s": 193.8868408203125}
+{"event": "train", "step": 7540, "loss": 3.493013119697571, "learning_rate": 0.00019420368446271936, "elapsed_s": 194.36360692977905}
+{"event": "train", "step": 7560, "loss": 3.448889064788818, "learning_rate": 0.000192980725928124, "elapsed_s": 194.81853795051575}
+{"event": "train", "step": 7580, "loss": 3.4463623642921446, "learning_rate": 0.00019175920505826044, "elapsed_s": 195.27398681640625}
+{"event": "train", "step": 7600, "loss": 3.4481689095497132, "learning_rate": 0.00019053915265214376, "elapsed_s": 195.7404341697693}
+{"event": "train", "step": 7620, "loss": 3.480922508239746, "learning_rate": 0.00018932059947176345, "elapsed_s": 196.19761562347412}
+{"event": "train", "step": 7640, "loss": 3.4478603839874267, "learning_rate": 0.00018810357624130813, "elapsed_s": 196.6537230014801}
+{"event": "train", "step": 7660, "loss": 3.472864592075348, "learning_rate": 0.0001868881136463908, "elapsed_s": 197.1158254146576}
+{"event": "train", "step": 7680, "loss": 3.4266767740249633, "learning_rate": 0.0001856742423332751, "elapsed_s": 197.57706785202026}
+{"event": "train", "step": 7700, "loss": 3.446815848350525, "learning_rate": 0.00018446199290810263, "elapsed_s": 198.03849005699158}
+{"event": "train", "step": 7720, "loss": 3.46683851480484, "learning_rate": 0.00018325139593612118, "elapsed_s": 198.49823307991028}
+{"event": "train", "step": 7740, "loss": 3.4551758646965025, "learning_rate": 0.00018204248194091428, "elapsed_s": 198.94707703590393}
+{"event": "train", "step": 7760, "loss": 3.3354167222976683, "learning_rate": 0.00018083528140363142, "elapsed_s": 199.43450546264648}
+{"event": "train", "step": 7780, "loss": 3.3744060039520263, "learning_rate": 0.00017962982476221946, "elapsed_s": 199.89540886878967}
+{"event": "train", "step": 7800, "loss": 3.3653141140937803, "learning_rate": 0.0001784261424106552, "elapsed_s": 200.3528003692627}
+{"event": "train", "step": 7820, "loss": 3.3870612382888794, "learning_rate": 0.0001772242646981794, "elapsed_s": 200.80805802345276}
+{"event": "train", "step": 7840, "loss": 3.38838267326355, "learning_rate": 0.00017602422192853094, "elapsed_s": 201.26404476165771}
+{"event": "train", "step": 7860, "loss": 3.391267251968384, "learning_rate": 0.00017482604435918315, "elapsed_s": 201.73456192016602}
+{"event": "train", "step": 7880, "loss": 3.374438762664795, "learning_rate": 0.00017362976220058093, "elapsed_s": 202.19812726974487}
+{"event": "train", "step": 7900, "loss": 3.391533851623535, "learning_rate": 0.0001724354056153789, "elapsed_s": 202.66426730155945}
+{"event": "train", "step": 7920, "loss": 3.3799129843711855, "learning_rate": 0.00017124300471768085, "elapsed_s": 203.12048935890198}
+{"event": "train", "step": 7940, "loss": 3.4042389154434205, "learning_rate": 0.00017005258957228047, "elapsed_s": 203.58196997642517}
+{"event": "train", "step": 7960, "loss": 3.381893587112427, "learning_rate": 0.00016886419019390364, "elapsed_s": 204.03626608848572}
+{"event": "train", "step": 7980, "loss": 3.4164822578430174, "learning_rate": 0.00016767783654645114, "elapsed_s": 204.48878407478333}
+{"event": "train", "step": 8000, "loss": 3.3934492468833923, "learning_rate": 0.00016649355854224336, "elapsed_s": 204.94297051429749}
+{"event": "validation", "step": 8000, "loss": 3.945939004421234, "recall_at_1": 0.019}
+{"event": "train", "step": 8020, "loss": 3.40130375623703, "learning_rate": 0.00016531138604126616, "elapsed_s": 205.68800020217896}
+{"event": "train", "step": 8040, "loss": 3.38695513010025, "learning_rate": 0.0001641313488504181, "elapsed_s": 206.1407880783081}
+{"event": "train", "step": 8060, "loss": 3.423900008201599, "learning_rate": 0.00016295347672275862, "elapsed_s": 206.60172033309937}
+{"event": "train", "step": 8080, "loss": 3.3947692036628725, "learning_rate": 0.00016177779935675792, "elapsed_s": 207.06409120559692}
+{"event": "train", "step": 8100, "loss": 3.4254462480545045, "learning_rate": 0.00016060434639554835, "elapsed_s": 207.5263078212738}
+{"event": "train", "step": 8120, "loss": 3.419986200332642, "learning_rate": 0.00015943314742617687, "elapsed_s": 207.98305654525757}
+{"event": "train", "step": 8140, "loss": 3.4377137660980224, "learning_rate": 0.00015826423197885895, "elapsed_s": 208.44074511528015}
+{"event": "train", "step": 8160, "loss": 3.3828097105026247, "learning_rate": 0.0001570976295262341, "elapsed_s": 208.8980438709259}
+{"event": "train", "step": 8180, "loss": 3.3776392102241517, "learning_rate": 0.00015593336948262296, "elapsed_s": 209.35953783988953}
+{"event": "train", "step": 8200, "loss": 3.320758783817291, "learning_rate": 0.00015477148120328525, "elapsed_s": 209.81989884376526}
+{"event": "train", "step": 8220, "loss": 3.3208076477050783, "learning_rate": 0.00015361199398367976, "elapsed_s": 210.278653383255}
+{"event": "train", "step": 8240, "loss": 3.305421733856201, "learning_rate": 0.00015245493705872621, "elapsed_s": 210.73888850212097}
+{"event": "train", "step": 8260, "loss": 3.348253917694092, "learning_rate": 0.0001513003396020672, "elapsed_s": 211.19864130020142}
+{"event": "train", "step": 8280, "loss": 3.3038049221038817, "learning_rate": 0.00015014823072533327, "elapsed_s": 211.66974544525146}
+{"event": "train", "step": 8300, "loss": 3.3276920437812807, "learning_rate": 0.00014899863947740878, "elapsed_s": 212.12671613693237}
+{"event": "train", "step": 8320, "loss": 3.306189274787903, "learning_rate": 0.00014785159484369964, "elapsed_s": 212.586030960083}
+{"event": "train", "step": 8340, "loss": 3.342453420162201, "learning_rate": 0.00014670712574540198, "elapsed_s": 213.04079365730286}
+{"event": "train", "step": 8360, "loss": 3.355472409725189, "learning_rate": 0.00014556526103877326, "elapsed_s": 213.49640274047852}
+{"event": "train", "step": 8380, "loss": 3.2991963386535645, "learning_rate": 0.00014442602951440507, "elapsed_s": 213.95682883262634}
+{"event": "train", "step": 8400, "loss": 3.3476312875747682, "learning_rate": 0.00014328945989649638, "elapsed_s": 214.41112875938416}
+{"event": "train", "step": 8420, "loss": 3.34841765165329, "learning_rate": 0.00014215558084213, "elapsed_s": 214.8651201725006}
+{"event": "train", "step": 8440, "loss": 3.3185312509536744, "learning_rate": 0.00014102442094054968, "elapsed_s": 215.3181140422821}
+{"event": "train", "step": 8460, "loss": 3.3097389221191404, "learning_rate": 0.00013989600871243938, "elapsed_s": 215.7706537246704}
+{"event": "train", "step": 8480, "loss": 3.3470398068428038, "learning_rate": 0.0001387703726092042, "elapsed_s": 216.22197890281677}
+{"event": "train", "step": 8500, "loss": 3.3698921680450438, "learning_rate": 0.00013764754101225287, "elapsed_s": 216.6742227077484}
+{"event": "train", "step": 8520, "loss": 3.3683993339538576, "learning_rate": 0.00013652754223228243, "elapsed_s": 217.1309895515442}
+{"event": "train", "step": 8540, "loss": 3.3304752469062806, "learning_rate": 0.00013541040450856402, "elapsed_s": 217.5857973098755}
+{"event": "train", "step": 8560, "loss": 3.3554144382476805, "learning_rate": 0.0001342961560082312, "elapsed_s": 218.04780983924866}
+{"event": "train", "step": 8580, "loss": 3.3931713104248047, "learning_rate": 0.00013318482482556958, "elapsed_s": 218.5069010257721}
+{"event": "train", "step": 8600, "loss": 3.354724872112274, "learning_rate": 0.00013207643898130854, "elapsed_s": 218.95037698745728}
+{"event": "train", "step": 8620, "loss": 3.267058491706848, "learning_rate": 0.0001309710264219146, "elapsed_s": 219.42769718170166}
+{"event": "train", "step": 8640, "loss": 3.277419400215149, "learning_rate": 0.00012986861501888686, "elapsed_s": 219.8794605731964}
+{"event": "train", "step": 8660, "loss": 3.289547324180603, "learning_rate": 0.00012876923256805456, "elapsed_s": 220.33992624282837}
+{"event": "train", "step": 8680, "loss": 3.2551644802093507, "learning_rate": 0.0001276729067888757, "elapsed_s": 220.79437851905823}
+{"event": "train", "step": 8700, "loss": 3.2746483087539673, "learning_rate": 0.00012657966532373847, "elapsed_s": 221.2544767856598}
+{"event": "train", "step": 8720, "loss": 3.29923335313797, "learning_rate": 0.00012548953573726413, "elapsed_s": 221.71745681762695}
+{"event": "train", "step": 8740, "loss": 3.274812090396881, "learning_rate": 0.00012440254551561217, "elapsed_s": 222.18428540229797}
+{"event": "train", "step": 8760, "loss": 3.3046647548675536, "learning_rate": 0.00012331872206578715, "elapsed_s": 222.64329528808594}
+{"event": "train", "step": 8780, "loss": 3.282978367805481, "learning_rate": 0.0001222380927149477, "elapsed_s": 223.10069584846497}
+{"event": "train", "step": 8800, "loss": 3.2895207524299623, "learning_rate": 0.00012116068470971763, "elapsed_s": 223.5562059879303}
+{"event": "train", "step": 8820, "loss": 3.2662514805793763, "learning_rate": 0.00012008652521549877, "elapsed_s": 224.01643800735474}
+{"event": "train", "step": 8840, "loss": 3.303015613555908, "learning_rate": 0.00011901564131578598, "elapsed_s": 224.47238564491272}
+{"event": "train", "step": 8860, "loss": 3.2717440128326416, "learning_rate": 0.00011794806001148442, "elapsed_s": 224.92976307868958}
+{"event": "train", "step": 8880, "loss": 3.3007007598876954, "learning_rate": 0.00011688380822022906, "elapsed_s": 225.3840835094452}
+{"event": "train", "step": 8900, "loss": 3.273569416999817, "learning_rate": 0.00011582291277570508, "elapsed_s": 225.83685398101807}
+{"event": "train", "step": 8920, "loss": 3.2740875124931335, "learning_rate": 0.00011476540042697217, "elapsed_s": 226.2892246246338}
+{"event": "train", "step": 8940, "loss": 3.2977951407432555, "learning_rate": 0.00011371129783778994, "elapsed_s": 226.74323105812073}
+{"event": "train", "step": 8960, "loss": 3.2869706034660338, "learning_rate": 0.0001126606315859452, "elapsed_s": 227.20814847946167}
+{"event": "train", "step": 8980, "loss": 3.2907532095909118, "learning_rate": 0.00011161342816258224, "elapsed_s": 227.81236505508423}
+{"event": "train", "step": 9000, "loss": 3.3052882313728333, "learning_rate": 0.00011056971397153462, "elapsed_s": 228.27330017089844}
+{"event": "validation", "step": 9000, "loss": 3.956260085105896, "recall_at_1": 0.017}
+{"event": "train", "step": 9020, "loss": 3.305304527282715, "learning_rate": 0.00010952951532865987, "elapsed_s": 229.065687417984}
+{"event": "train", "step": 9040, "loss": 3.212058162689209, "learning_rate": 0.00010849285846117543, "elapsed_s": 229.53025579452515}
+{"event": "train", "step": 9060, "loss": 3.2319684743881227, "learning_rate": 0.00010745976950699765, "elapsed_s": 229.9831256866455}
+{"event": "train", "step": 9080, "loss": 3.235683834552765, "learning_rate": 0.00010643027451408274, "elapsed_s": 230.43955326080322}
+{"event": "train", "step": 9100, "loss": 3.202328622341156, "learning_rate": 0.00010540439943977, "elapsed_s": 230.8926727771759}
+{"event": "train", "step": 9120, "loss": 3.2156752705574037, "learning_rate": 0.00010438217015012735, "elapsed_s": 231.34818387031555}
+{"event": "train", "step": 9140, "loss": 3.227373206615448, "learning_rate": 0.00010336361241929893, "elapsed_s": 231.80358982086182}
+{"event": "train", "step": 9160, "loss": 3.2351808190345763, "learning_rate": 0.00010234875192885576, "elapsed_s": 232.2686095237732}
+{"event": "train", "step": 9180, "loss": 3.2246779799461365, "learning_rate": 0.00010133761426714771, "elapsed_s": 232.7283158302307}
+{"event": "train", "step": 9200, "loss": 3.199734091758728, "learning_rate": 0.00010033022492865846, "elapsed_s": 233.20059370994568}
+{"event": "train", "step": 9220, "loss": 3.230129563808441, "learning_rate": 9.932660931336279e-05, "elapsed_s": 233.67618703842163}
+{"event": "train", "step": 9240, "loss": 3.246441924571991, "learning_rate": 9.832679272608607e-05, "elapsed_s": 234.13713192939758}
+{"event": "train", "step": 9260, "loss": 3.2459362268447878, "learning_rate": 9.733080037586636e-05, "elapsed_s": 234.5976903438568}
+{"event": "train", "step": 9280, "loss": 3.2356390237808226, "learning_rate": 9.633865737531847e-05, "elapsed_s": 235.05509495735168}
+{"event": "train", "step": 9300, "loss": 3.253940498828888, "learning_rate": 9.535038874000134e-05, "elapsed_s": 235.51016855239868}
+{"event": "train", "step": 9320, "loss": 3.225062882900238, "learning_rate": 9.436601938778672e-05, "elapsed_s": 235.96618008613586}
+{"event": "train", "step": 9340, "loss": 3.244731402397156, "learning_rate": 9.338557413823123e-05, "elapsed_s": 236.4265410900116}
+{"event": "train", "step": 9360, "loss": 3.258590269088745, "learning_rate": 9.240907771195046e-05, "elapsed_s": 236.9083023071289}
+{"event": "train", "step": 9380, "loss": 3.2337982296943664, "learning_rate": 9.143655472999571e-05, "elapsed_s": 237.37187480926514}
+{"event": "train", "step": 9400, "loss": 3.234100413322449, "learning_rate": 9.046802971323323e-05, "elapsed_s": 237.8319902420044}
+{"event": "train", "step": 9420, "loss": 3.2520681262016295, "learning_rate": 8.950352708172583e-05, "elapsed_s": 238.29250383377075}
+{"event": "train", "step": 9440, "loss": 3.2307455062866213, "learning_rate": 8.85430711541175e-05, "elapsed_s": 238.7546193599701}
+{"event": "train", "step": 9460, "loss": 3.2404507756233216, "learning_rate": 8.758668614701973e-05, "elapsed_s": 239.20540404319763}
+{"event": "train", "step": 9480, "loss": 3.1887077927589416, "learning_rate": 8.663439617440133e-05, "elapsed_s": 239.68956351280212}
+{"event": "train", "step": 9500, "loss": 3.1787684082984926, "learning_rate": 8.568622524698025e-05, "elapsed_s": 240.15136456489563}
+{"event": "train", "step": 9520, "loss": 3.131831443309784, "learning_rate": 8.474219727161823e-05, "elapsed_s": 240.61002731323242}
+{"event": "train", "step": 9540, "loss": 3.1589084386825563, "learning_rate": 8.3802336050718e-05, "elapsed_s": 241.06824111938477}
+{"event": "train", "step": 9560, "loss": 3.1678595423698424, "learning_rate": 8.286666528162312e-05, "elapsed_s": 241.52577567100525}
+{"event": "train", "step": 9580, "loss": 3.1866217494010924, "learning_rate": 8.193520855602071e-05, "elapsed_s": 241.98672938346863}
+{"event": "train", "step": 9600, "loss": 3.191297376155853, "learning_rate": 8.100798935934622e-05, "elapsed_s": 242.45767545700073}
+{"event": "train", "step": 9620, "loss": 3.2053757548332213, "learning_rate": 8.008503107019152e-05, "elapsed_s": 242.91777229309082}
+{"event": "train", "step": 9640, "loss": 3.216445338726044, "learning_rate": 7.916635695971539e-05, "elapsed_s": 243.39964079856873}
+{"event": "train", "step": 9660, "loss": 3.1828588604927064, "learning_rate": 7.825199019105677e-05, "elapsed_s": 243.8609869480133}
+{"event": "train", "step": 9680, "loss": 3.214737629890442, "learning_rate": 7.73419538187508e-05, "elapsed_s": 244.32862496376038}
+{"event": "train", "step": 9700, "loss": 3.1826008319854737, "learning_rate": 7.643627078814733e-05, "elapsed_s": 244.78986263275146}
+{"event": "train", "step": 9720, "loss": 3.1660401940345766, "learning_rate": 7.553496393483278e-05, "elapsed_s": 245.2591414451599}
+{"event": "train", "step": 9740, "loss": 3.206581676006317, "learning_rate": 7.463805598405384e-05, "elapsed_s": 245.72414875030518}
+{"event": "train", "step": 9760, "loss": 3.1558589577674865, "learning_rate": 7.374556955014491e-05, "elapsed_s": 246.19106912612915}
+{"event": "train", "step": 9780, "loss": 3.20000159740448, "learning_rate": 7.285752713595762e-05, "elapsed_s": 246.66786241531372}
+{"event": "train", "step": 9800, "loss": 3.197562110424042, "learning_rate": 7.197395113229388e-05, "elapsed_s": 247.1261920928955}
+{"event": "train", "step": 9820, "loss": 3.169292390346527, "learning_rate": 7.109486381734062e-05, "elapsed_s": 247.58604192733765}
+{"event": "train", "step": 9840, "loss": 3.176116919517517, "learning_rate": 7.022028735610869e-05, "elapsed_s": 248.04646062850952}
+{"event": "train", "step": 9860, "loss": 3.179330515861511, "learning_rate": 6.935024379987401e-05, "elapsed_s": 248.50842809677124}
+{"event": "train", "step": 9880, "loss": 3.183117139339447, "learning_rate": 6.848475508562103e-05, "elapsed_s": 248.96738028526306}
+{"event": "train", "step": 9900, "loss": 3.1383650422096254, "learning_rate": 6.762384303549005e-05, "elapsed_s": 249.43608450889587}
+{"event": "train", "step": 9920, "loss": 3.131430685520172, "learning_rate": 6.676752935622694e-05, "elapsed_s": 249.91747856140137}
+{"event": "train", "step": 9940, "loss": 3.1481836915016173, "learning_rate": 6.591583563863588e-05, "elapsed_s": 250.37591361999512}
+{"event": "train", "step": 9960, "loss": 3.1337304592132567, "learning_rate": 6.506878335703483e-05, "elapsed_s": 250.83943390846252}
+{"event": "train", "step": 9980, "loss": 3.147600495815277, "learning_rate": 6.422639386871393e-05, "elapsed_s": 251.29616594314575}
+{"event": "train", "step": 10000, "loss": 3.1395550966262817, "learning_rate": 6.338868841339763e-05, "elapsed_s": 251.75235176086426}
+{"event": "validation", "step": 10000, "loss": 4.000731736421585, "recall_at_1": 0.0185}
+{"event": "train", "step": 10020, "loss": 3.139786720275879, "learning_rate": 6.255568811270848e-05, "elapsed_s": 252.47329831123352}
+{"event": "train", "step": 10040, "loss": 3.107504725456238, "learning_rate": 6.172741396963488e-05, "elapsed_s": 252.95637607574463}
+{"event": "train", "step": 10060, "loss": 3.1710439324378967, "learning_rate": 6.090388686800147e-05, "elapsed_s": 253.42460441589355}
+{"event": "train", "step": 10080, "loss": 3.1660329341888427, "learning_rate": 6.0085127571942815e-05, "elapsed_s": 253.89220547676086}
+{"event": "train", "step": 10100, "loss": 3.107550299167633, "learning_rate": 5.927115672537936e-05, "elapsed_s": 254.35504937171936}
+{"event": "train", "step": 10120, "loss": 3.1543678522109984, "learning_rate": 5.8461994851497306e-05, "elapsed_s": 254.81941175460815}
+{"event": "train", "step": 10140, "loss": 3.1170974373817444, "learning_rate": 5.765766235223108e-05, "elapsed_s": 255.28007173538208}
+{"event": "train", "step": 10160, "loss": 3.130656909942627, "learning_rate": 5.68581795077488e-05, "elapsed_s": 255.73681116104126}
+{"event": "train", "step": 10180, "loss": 3.165736699104309, "learning_rate": 5.60635664759411e-05, "elapsed_s": 256.1957187652588}
+{"event": "train", "step": 10200, "loss": 3.127567195892334, "learning_rate": 5.5273843291912704e-05, "elapsed_s": 256.64907026290894}
+{"event": "train", "step": 10220, "loss": 3.1493678212165834, "learning_rate": 5.448902986747756e-05, "elapsed_s": 257.1045734882355}
+{"event": "train", "step": 10240, "loss": 3.1683237195014953, "learning_rate": 5.37091459906564e-05, "elapsed_s": 257.56819248199463}
+{"event": "train", "step": 10260, "loss": 3.133812892436981, "learning_rate": 5.293421132517809e-05, "elapsed_s": 258.031281709671}
+{"event": "train", "step": 10280, "loss": 3.146796178817749, "learning_rate": 5.216424540998374e-05, "elapsed_s": 258.49420285224915}
+{"event": "train", "step": 10300, "loss": 3.128574812412262, "learning_rate": 5.139926765873404e-05, "elapsed_s": 258.959899187088}
+{"event": "train", "step": 10320, "loss": 3.1122903704643248, "learning_rate": 5.063929735931985e-05, "elapsed_s": 259.4187400341034}
+{"event": "train", "step": 10340, "loss": 3.0808979153633116, "learning_rate": 4.988435367337571e-05, "elapsed_s": 259.89547061920166}
+{"event": "train", "step": 10360, "loss": 3.0823850989341737, "learning_rate": 4.913445563579705e-05, "elapsed_s": 260.347838640213}
+{"event": "train", "step": 10380, "loss": 3.098862648010254, "learning_rate": 4.8389622154259874e-05, "elapsed_s": 260.8026201725006}
+{"event": "train", "step": 10400, "loss": 3.1101354241371153, "learning_rate": 4.764987200874413e-05, "elapsed_s": 261.25787329673767}
+{"event": "train", "step": 10420, "loss": 3.0533060431480408, "learning_rate": 4.691522385106034e-05, "elapsed_s": 261.7140519618988}
+{"event": "train", "step": 10440, "loss": 3.119806146621704, "learning_rate": 4.618569620437921e-05, "elapsed_s": 262.1674003601074}
+{"event": "train", "step": 10460, "loss": 3.093064069747925, "learning_rate": 4.546130746276461e-05, "elapsed_s": 262.6425988674164}
+{"event": "train", "step": 10480, "loss": 3.067823588848114, "learning_rate": 4.474207589070969e-05, "elapsed_s": 263.103164434433}
+{"event": "train", "step": 10500, "loss": 3.102240788936615, "learning_rate": 4.402801962267669e-05, "elapsed_s": 263.56722021102905}
+{"event": "train", "step": 10520, "loss": 3.124609410762787, "learning_rate": 4.3319156662639306e-05, "elapsed_s": 264.0295090675354}
+{"event": "train", "step": 10540, "loss": 3.123835802078247, "learning_rate": 4.261550488362895e-05, "elapsed_s": 264.504647731781}
+{"event": "train", "step": 10560, "loss": 3.0786010384559632, "learning_rate": 4.191708202728411e-05, "elapsed_s": 264.96406745910645}
+{"event": "train", "step": 10580, "loss": 3.122776710987091, "learning_rate": 4.122390570340292e-05, "elapsed_s": 265.4244441986084}
+{"event": "train", "step": 10600, "loss": 3.13657408952713, "learning_rate": 4.0535993389499285e-05, "elapsed_s": 265.8811705112457}
+{"event": "train", "step": 10620, "loss": 3.1353241086006163, "learning_rate": 3.9853362430362055e-05, "elapsed_s": 266.33947944641113}
+{"event": "train", "step": 10640, "loss": 3.092473089694977, "learning_rate": 3.917603003761797e-05, "elapsed_s": 266.7960829734802}
+{"event": "train", "step": 10660, "loss": 3.1207069396972655, "learning_rate": 3.850401328929726e-05, "elapsed_s": 267.25946521759033}
+{"event": "train", "step": 10680, "loss": 3.07033885717392, "learning_rate": 3.783732912940341e-05, "elapsed_s": 267.8326539993286}
+{"event": "train", "step": 10700, "loss": 3.1097807049751283, "learning_rate": 3.717599436748595e-05, "elapsed_s": 268.29342103004456}
+{"event": "train", "step": 10720, "loss": 3.0855018138885497, "learning_rate": 3.652002567821619e-05, "elapsed_s": 268.75547432899475}
+{"event": "train", "step": 10740, "loss": 3.0615798234939575, "learning_rate": 3.586943960096731e-05, "elapsed_s": 269.2188346385956}
+{"event": "train", "step": 10760, "loss": 3.0803686499595644, "learning_rate": 3.522425253939707e-05, "elapsed_s": 269.6843602657318}
+{"event": "train", "step": 10780, "loss": 3.0572709918022154, "learning_rate": 3.4584480761034376e-05, "elapsed_s": 270.1764371395111}
+{"event": "train", "step": 10800, "loss": 3.0758774995803835, "learning_rate": 3.395014039686883e-05, "elapsed_s": 270.67268681526184}
+{"event": "train", "step": 10820, "loss": 3.0631178617477417, "learning_rate": 3.3321247440944336e-05, "elapsed_s": 271.12308263778687}
+{"event": "train", "step": 10840, "loss": 3.0043989181518556, "learning_rate": 3.269781774995551e-05, "elapsed_s": 271.57481598854065}
+{"event": "train", "step": 10860, "loss": 3.0368908047676086, "learning_rate": 3.2079867042848375e-05, "elapsed_s": 272.0285792350769}
+{"event": "train", "step": 10880, "loss": 3.063518500328064, "learning_rate": 3.146741090042332e-05, "elapsed_s": 272.4815135002136}
+{"event": "train", "step": 10900, "loss": 3.042289447784424, "learning_rate": 3.0860464764942805e-05, "elapsed_s": 272.9631109237671}
+{"event": "train", "step": 10920, "loss": 3.0722107887268066, "learning_rate": 3.0259043939741964e-05, "elapsed_s": 273.45552110671997}
+{"event": "train", "step": 10940, "loss": 3.07947313785553, "learning_rate": 2.9663163588842414e-05, "elapsed_s": 273.9447247982025}
+{"event": "train", "step": 10960, "loss": 3.0826860547065733, "learning_rate": 2.907283873657024e-05, "elapsed_s": 274.41059827804565}
+{"event": "train", "step": 10980, "loss": 3.0808752298355104, "learning_rate": 2.8488084267176945e-05, "elapsed_s": 274.9052393436432}
+{"event": "train", "step": 11000, "loss": 3.0654750227928163, "learning_rate": 2.7908914924464486e-05, "elapsed_s": 275.3815379142761}
+{"event": "validation", "step": 11000, "loss": 4.009405940771103, "recall_at_1": 0.0185}
+{"event": "train", "step": 11020, "loss": 3.0683713793754577, "learning_rate": 2.7335345311413183e-05, "elapsed_s": 276.1634223461151}
+{"event": "train", "step": 11040, "loss": 3.07088098526001, "learning_rate": 2.6767389889813502e-05, "elapsed_s": 276.6182451248169}
+{"event": "train", "step": 11060, "loss": 3.0660889506340028, "learning_rate": 2.6205062979901968e-05, "elapsed_s": 277.07254219055176}
+{"event": "train", "step": 11080, "loss": 3.081027698516846, "learning_rate": 2.564837875999937e-05, "elapsed_s": 277.528028011322}
+{"event": "train", "step": 11100, "loss": 3.0956427097320556, "learning_rate": 2.5097351266153785e-05, "elapsed_s": 277.98328256607056}
+{"event": "train", "step": 11120, "loss": 3.0462390661239622, "learning_rate": 2.4551994391786404e-05, "elapsed_s": 278.4443852901459}
+{"event": "train", "step": 11140, "loss": 3.080029833316803, "learning_rate": 2.401232188734151e-05, "elapsed_s": 278.90078020095825}
+{"event": "train", "step": 11160, "loss": 3.0709501981735228, "learning_rate": 2.347834735993942e-05, "elapsed_s": 279.366126537323}
+{"event": "train", "step": 11180, "loss": 3.062103283405304, "learning_rate": 2.2950084273033632e-05, "elapsed_s": 279.809974193573}
+{"event": "train", "step": 11200, "loss": 3.0398420810699465, "learning_rate": 2.2427545946071354e-05, "elapsed_s": 280.2896761894226}
+{"event": "train", "step": 11220, "loss": 3.0240533471107485, "learning_rate": 2.1910745554157563e-05, "elapsed_s": 280.74650716781616}
+{"event": "train", "step": 11240, "loss": 3.030126214027405, "learning_rate": 2.1399696127722967e-05, "elapsed_s": 281.20072078704834}
+{"event": "train", "step": 11260, "loss": 3.0076622009277343, "learning_rate": 2.0894410552195237e-05, "elapsed_s": 281.65886545181274}
+{"event": "train", "step": 11280, "loss": 3.0418915510177613, "learning_rate": 2.039490156767443e-05, "elapsed_s": 282.116623878479}
+{"event": "train", "step": 11300, "loss": 3.0445916652679443, "learning_rate": 1.990118176861147e-05, "elapsed_s": 282.57982301712036}
+{"event": "train", "step": 11320, "loss": 3.0424877643585204, "learning_rate": 1.9413263603490688e-05, "elapsed_s": 283.0460116863251}
+{"event": "train", "step": 11340, "loss": 3.0555174708366395, "learning_rate": 1.893115937451609e-05, "elapsed_s": 283.509578704834}
+{"event": "train", "step": 11360, "loss": 3.024147295951843, "learning_rate": 1.8454881237300935e-05, "elapsed_s": 283.9724631309509}
+{"event": "train", "step": 11380, "loss": 3.0416974067687987, "learning_rate": 1.798444120056142e-05, "elapsed_s": 284.42532992362976}
+{"event": "train", "step": 11400, "loss": 3.025550961494446, "learning_rate": 1.7519851125813863e-05, "elapsed_s": 284.87810254096985}
+{"event": "train", "step": 11420, "loss": 3.0249926090240478, "learning_rate": 1.7061122727075662e-05, "elapsed_s": 285.34083366394043}
+{"event": "train", "step": 11440, "loss": 3.0398239970207213, "learning_rate": 1.660826757056977e-05, "elapsed_s": 285.7935173511505}
+{"event": "train", "step": 11460, "loss": 3.0303362727165224, "learning_rate": 1.6161297074433284e-05, "elapsed_s": 286.24603629112244}
+{"event": "train", "step": 11480, "loss": 3.040290820598602, "learning_rate": 1.5720222508429442e-05, "elapsed_s": 286.69573616981506}
+{"event": "train", "step": 11500, "loss": 3.001080024242401, "learning_rate": 1.5285054993663506e-05, "elapsed_s": 287.14559674263}
+{"event": "train", "step": 11520, "loss": 3.0427719831466673, "learning_rate": 1.4855805502302255e-05, "elapsed_s": 287.59624552726746}
+{"event": "train", "step": 11540, "loss": 3.0471933603286745, "learning_rate": 1.4432484857297507e-05, "elapsed_s": 288.04579520225525}
+{"event": "train", "step": 11560, "loss": 3.0660608530044557, "learning_rate": 1.4015103732113177e-05, "elapsed_s": 288.49457359313965}
+{"event": "train", "step": 11580, "loss": 3.0254424571990968, "learning_rate": 1.3603672650456062e-05, "elapsed_s": 288.9455361366272}
+{"event": "train", "step": 11600, "loss": 3.0629366755485536, "learning_rate": 1.3198201986010572e-05, "elapsed_s": 289.3977324962616}
+{"event": "train", "step": 11620, "loss": 3.047807812690735, "learning_rate": 1.2798701962177223e-05, "elapsed_s": 289.86643052101135}
+{"event": "train", "step": 11640, "loss": 3.001612055301666, "learning_rate": 1.2405182651814755e-05, "elapsed_s": 290.32985830307007}
+{"event": "train", "step": 11660, "loss": 3.0193174481391907, "learning_rate": 1.201765397698626e-05, "elapsed_s": 290.7849054336548}
+{"event": "train", "step": 11680, "loss": 3.028712201118469, "learning_rate": 1.1636125708708895e-05, "elapsed_s": 291.2360625267029}
+{"event": "train", "step": 11700, "loss": 3.0144689798355104, "learning_rate": 1.1260607466707733e-05, "elapsed_s": 291.68977308273315}
+{"event": "train", "step": 11720, "loss": 3.026969647407532, "learning_rate": 1.0891108719172926e-05, "elapsed_s": 292.1395683288574}
+{"event": "train", "step": 11740, "loss": 3.012769675254822, "learning_rate": 1.0527638782521193e-05, "elapsed_s": 292.5896027088165}
+{"event": "train", "step": 11760, "loss": 3.0180457711219786, "learning_rate": 1.0170206821160905e-05, "elapsed_s": 293.03979897499084}
+{"event": "train", "step": 11780, "loss": 3.0410977363586427, "learning_rate": 9.818821847260822e-06, "elapsed_s": 293.4902937412262}
+{"event": "train", "step": 11800, "loss": 3.0125139117240907, "learning_rate": 9.473492720523113e-06, "elapsed_s": 293.9674401283264}
+{"event": "train", "step": 11820, "loss": 2.9848084092140197, "learning_rate": 9.134228147959778e-06, "elapsed_s": 294.425669670105}
+{"event": "train", "step": 11840, "loss": 3.025339591503143, "learning_rate": 8.801036683673325e-06, "elapsed_s": 294.88789534568787}
+{"event": "train", "step": 11860, "loss": 2.9946898341178896, "learning_rate": 8.473926728640834e-06, "elapsed_s": 295.363965511322}
+{"event": "train", "step": 11880, "loss": 3.0248006224632262, "learning_rate": 8.152906530502296e-06, "elapsed_s": 295.81732845306396}
+{"event": "train", "step": 11900, "loss": 3.0139110922813415, "learning_rate": 7.837984183352658e-06, "elapsed_s": 296.2663161754608}
+{"event": "train", "step": 11920, "loss": 3.0140116572380067, "learning_rate": 7.529167627537748e-06, "elapsed_s": 296.7145507335663}
+{"event": "train", "step": 11940, "loss": 3.02456419467926, "learning_rate": 7.2264646494538675e-06, "elapsed_s": 297.16279339790344}
+{"event": "train", "step": 11960, "loss": 3.0322821855545046, "learning_rate": 6.929882881351795e-06, "elapsed_s": 297.6105523109436}
+{"event": "train", "step": 11980, "loss": 3.026415002346039, "learning_rate": 6.6394298011441255e-06, "elapsed_s": 298.05919671058655}
+{"event": "train", "step": 12000, "loss": 3.0160702586174013, "learning_rate": 6.355112732216811e-06, "elapsed_s": 298.50777292251587}
+{"event": "validation", "step": 12000, "loss": 4.023470968008041, "recall_at_1": 0.0155}
+{"event": "train", "step": 12020, "loss": 3.0097503423690797, "learning_rate": 6.076938843244423e-06, "elapsed_s": 299.2639191150665}
+{"event": "train", "step": 12040, "loss": 2.9934877872467043, "learning_rate": 5.804915148009571e-06, "elapsed_s": 299.701299905777}
+{"event": "train", "step": 12060, "loss": 2.996999907493591, "learning_rate": 5.53904850522588e-06, "elapsed_s": 300.18824791908264}
+{"event": "train", "step": 12080, "loss": 2.9984724521636963, "learning_rate": 5.279345618365128e-06, "elapsed_s": 300.6498672962189}
+{"event": "train", "step": 12100, "loss": 3.0115355849266052, "learning_rate": 5.025813035488242e-06, "elapsed_s": 301.1040461063385}
+{"event": "train", "step": 12120, "loss": 3.0026384353637696, "learning_rate": 4.778457149080207e-06, "elapsed_s": 301.5540609359741}
+{"event": "train", "step": 12140, "loss": 3.009648072719574, "learning_rate": 4.537284195888891e-06, "elapsed_s": 302.00353598594666}
+{"event": "train", "step": 12160, "loss": 2.980654764175415, "learning_rate": 4.302300256767671e-06, "elapsed_s": 302.4524612426758}
+{"event": "train", "step": 12180, "loss": 3.0194852709770204, "learning_rate": 4.073511256522333e-06, "elapsed_s": 302.90070390701294}
+{"event": "train", "step": 12200, "loss": 2.987524890899658, "learning_rate": 3.850922963761522e-06, "elapsed_s": 303.3538887500763}
+{"event": "train", "step": 12220, "loss": 2.9751698970794678, "learning_rate": 3.6345409907512783e-06, "elapsed_s": 303.81431674957275}
+{"event": "train", "step": 12240, "loss": 2.9881492495536803, "learning_rate": 3.424370793273679e-06, "elapsed_s": 304.2742865085602}
+{"event": "train", "step": 12260, "loss": 2.984150755405426, "learning_rate": 3.220417670489112e-06, "elapsed_s": 304.73417043685913}
+{"event": "train", "step": 12280, "loss": 2.9934112191200257, "learning_rate": 3.022686764802801e-06, "elapsed_s": 305.19353699684143}
+{"event": "train", "step": 12300, "loss": 2.9708714485168457, "learning_rate": 2.831183061735049e-06, "elapsed_s": 305.6534788608551}
+{"event": "train", "step": 12320, "loss": 3.0274218916893005, "learning_rate": 2.6459113897956135e-06, "elapsed_s": 306.1092584133148}
+{"event": "train", "step": 12340, "loss": 3.03188716173172, "learning_rate": 2.466876420361863e-06, "elapsed_s": 306.5608892440796}
+{"event": "train", "step": 12360, "loss": 3.0136942505836486, "learning_rate": 2.294082667561148e-06, "elapsed_s": 307.01255679130554}
+{"event": "train", "step": 12380, "loss": 3.0083579063415526, "learning_rate": 2.127534488156807e-06, "elapsed_s": 307.5729646682739}
+{"event": "train", "step": 12400, "loss": 2.9925155639648438, "learning_rate": 1.967236081438395e-06, "elapsed_s": 308.0253872871399}
+{"event": "train", "step": 12420, "loss": 2.9811888337135315, "learning_rate": 1.8131914891158786e-06, "elapsed_s": 308.47886991500854}
+{"event": "train", "step": 12440, "loss": 3.037796342372894, "learning_rate": 1.6654045952175789e-06, "elapsed_s": 308.93265199661255}
+{"event": "train", "step": 12460, "loss": 2.974989318847656, "learning_rate": 1.523879125992389e-06, "elapsed_s": 309.38893389701843}
+{"event": "train", "step": 12480, "loss": 2.993820309638977, "learning_rate": 1.3886186498157094e-06, "elapsed_s": 309.85213804244995}
+{"event": "train", "step": 12500, "loss": 3.0071717381477354, "learning_rate": 1.2596265770995208e-06, "elapsed_s": 310.3067443370819}
+{"event": "train", "step": 12520, "loss": 3.01785728931427, "learning_rate": 1.1369061602064524e-06, "elapsed_s": 310.76067566871643}
+{"event": "train", "step": 12540, "loss": 2.9834540247917176, "learning_rate": 1.0204604933676531e-06, "elapsed_s": 311.2159867286682}
+{"event": "train", "step": 12560, "loss": 2.985476243495941, "learning_rate": 9.102925126048267e-07, "elapsed_s": 311.6680862903595}
+{"event": "train", "step": 12580, "loss": 2.9917554020881654, "learning_rate": 8.06404995656318e-07, "elapsed_s": 312.1197700500488}
+{"event": "train", "step": 12600, "loss": 3.031731462478638, "learning_rate": 7.088005619068916e-07, "elapsed_s": 312.57270884513855}
+{"event": "train", "step": 12620, "loss": 3.0241649866104128, "learning_rate": 6.174816723217847e-07, "elapsed_s": 313.0248029232025}
+{"event": "train", "step": 12640, "loss": 2.992753744125366, "learning_rate": 5.324506293846732e-07, "elapsed_s": 313.4876971244812}
+{"event": "train", "step": 12660, "loss": 2.9793124318122866, "learning_rate": 4.537095770395516e-07, "elapsed_s": 313.94764471054077}
+{"event": "train", "step": 12680, "loss": 2.9921557188034056, "learning_rate": 3.8126050063680416e-07, "elapsed_s": 314.40341997146606}
+{"event": "train", "step": 12700, "loss": 2.9569194197654722, "learning_rate": 3.151052268829113e-07, "elapsed_s": 314.86352372169495}
+{"event": "train", "step": 12720, "loss": 2.980341148376465, "learning_rate": 2.552454237946533e-07, "elapsed_s": 315.3205101490021}
+{"event": "train", "step": 12740, "loss": 3.0332400798797607, "learning_rate": 2.0168260065689393e-07, "elapsed_s": 315.7773222923279}
+{"event": "train", "step": 12760, "loss": 3.0049991488456724, "learning_rate": 1.54418107984583e-07, "elapsed_s": 316.23370456695557}
+{"event": "train", "step": 12780, "loss": 2.9720236897468566, "learning_rate": 1.1345313748870045e-07, "elapsed_s": 316.688321352005}
+{"event": "train", "step": 12800, "loss": 2.98990284204483, "learning_rate": 7.878872204616916e-08, "elapsed_s": 317.1407313346863}
+{"event": "train", "step": 12820, "loss": 2.986050033569336, "learning_rate": 5.0425735673848004e-08, "elapsed_s": 317.59672117233276}
+{"event": "train", "step": 12840, "loss": 2.9973851084709167, "learning_rate": 2.836489350654947e-08, "elapsed_s": 318.05614042282104}
+{"event": "train", "step": 12860, "loss": 3.0079408645629884, "learning_rate": 1.2606751778859727e-08, "elapsed_s": 318.5093991756439}
+{"event": "train", "step": 12880, "loss": 3.0100274682044983, "learning_rate": 3.1517078112330844e-09, "elapsed_s": 318.9621183872223}
+{"event": "train", "step": 12900, "loss": 3.0249800086021423, "learning_rate": 0.0, "elapsed_s": 319.4003460407257}
+{"event": "final", "step": 12900, "train_loss": 2.947625160217285, "val_loss": 4.031911849975586, "val_recall_at_1": 0.0165}

paper_experiments/2026-09-13/checkpoints/seed44/best/meta.json

@@ -0,0 +1,5 @@
+{
+  "step": 4000,
+  "val_loss": 4.107629656791687,
+  "val_recall@1": 0.015
+}

paper_experiments/2026-09-13/checkpoints/seed44/best/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5b20bc048de14602a91a3ee8730ef0b059a2a091083b5bd631fef9a95b39fdf
+size 6622652

paper_experiments/2026-09-13/checkpoints/seed44/config.yaml

@@ -0,0 +1,59 @@
+data:
+  aligned_dir: /workspace/artifacts/galaxy-image-spectrum-clip-ee2e189/precomputed_features
+  shard_cache_size: 0
+  image:
+    bands:
+    - g
+    - r
+    - z
+    image_size: 224
+    arcsinh_scale: 0.1
+  spectrum:
+    wavelength_min: 3600.0
+    wavelength_max: 9800.0
+    n_bins: 1024
+image_encoder:
+  type: identity
+  output_dim: 1024
+spectrum_encoder:
+  hidden_channels:
+  - 64
+  - 128
+  - 256
+  kernel_size: 5
+  embedding_dim: 512
+  dropout: 0.0
+  augment_noise_std: 0.0
+  augment_mask_frac: 0.0
+projection:
+  shared_dim: 512
+  hidden_dim: 512
+  dropout: 0.0
+model:
+  temperature_init: 0.07
+  device: cuda
+  torch_dtype: bfloat16
+training:
+  output_dir: /workspace/runs/gpu/baseline-seed44/checkpoints
+  num_epochs: 30
+  per_device_batch_size: 256
+  gradient_accumulation_steps: 1
+  learning_rate: 0.0005
+  warmup_ratio: 0.03
+  weight_decay: 0.05
+  max_grad_norm: 1.0
+  bf16: true
+  logging_steps: 20
+  eval_steps: 1000
+  save_steps: 0
+  eval_num_samples: 2000
+  dataloader_num_workers: 0
+  seed: 44
+eval:
+  recall_ks:
+  - 1
+  - 5
+  - 10
+  probe_targets:
+  - redshift
+  metrics_dir: /workspace/runs/gpu/baseline-seed44/metrics

paper_experiments/2026-09-13/checkpoints/seed44/evaluation.json

@@ -0,0 +1,34 @@
+{
+  "retrieval": {
+    "n": 13762,
+    "recall@1": 0.002397907280918471,
+    "recall@5": 0.011335561691614591,
+    "recall@10": 0.021072518529283536,
+    "median_rank": 720.5,
+    "mrr": 0.011326815378060795,
+    "image_to_spectrum_recall@1": 0.002397907280918471,
+    "image_to_spectrum_recall@5": 0.011335561691614591,
+    "image_to_spectrum_recall@10": 0.021072518529283536,
+    "image_to_spectrum_median_rank": 720.5,
+    "image_to_spectrum_mrr": 0.011326815378060795,
+    "spectrum_to_image_recall@1": 0.0027612265659061183,
+    "spectrum_to_image_recall@5": 0.011117570120622002,
+    "spectrum_to_image_recall@10": 0.019401249818340356,
+    "spectrum_to_image_median_rank": 790.5,
+    "spectrum_to_image_mrr": 0.011135130944609599
+  },
+  "probe_redshift": {
+    "image": {
+      "r2": 0.6832029819488525,
+      "mse": 0.010979507118463516,
+      "n_train": 110098,
+      "n_test": 13762
+    },
+    "spectrum": {
+      "r2": 0.8226574063301086,
+      "mse": 0.006146314088255167,
+      "n_train": 110098,
+      "n_test": 13762
+    }
+  }
+}

paper_experiments/2026-09-13/checkpoints/seed44/final/meta.json

@@ -0,0 +1,6 @@
+{
+  "step": 12900,
+  "train_loss": 2.999325752258301,
+  "val_loss": 4.116906225681305,
+  "val_recall@1": 0.0115
+}

paper_experiments/2026-09-13/checkpoints/seed44/final/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f81347a7cfb6694850d461140d58cd586d94d66e585c2836b08b9b9fb4009f66
+size 6622652

paper_experiments/2026-09-13/checkpoints/seed44/history.jsonl

@@ -0,0 +1,658 @@
+{"event": "train", "step": 20, "loss": 5.521001505851745, "learning_rate": 2.5839793281653746e-05, "elapsed_s": 9.41556191444397}
+{"event": "train", "step": 40, "loss": 5.334772419929505, "learning_rate": 5.167958656330749e-05, "elapsed_s": 9.871832132339478}
+{"event": "train", "step": 60, "loss": 5.104804539680481, "learning_rate": 7.751937984496124e-05, "elapsed_s": 10.322341918945312}
+{"event": "train", "step": 80, "loss": 4.960092806816101, "learning_rate": 0.00010335917312661498, "elapsed_s": 10.770281791687012}
+{"event": "train", "step": 100, "loss": 4.876599931716919, "learning_rate": 0.00012919896640826872, "elapsed_s": 11.235175609588623}
+{"event": "train", "step": 120, "loss": 4.819259142875671, "learning_rate": 0.0001550387596899225, "elapsed_s": 11.690858364105225}
+{"event": "train", "step": 140, "loss": 4.761411738395691, "learning_rate": 0.00018087855297157623, "elapsed_s": 12.144574165344238}
+{"event": "train", "step": 160, "loss": 4.747171926498413, "learning_rate": 0.00020671834625322997, "elapsed_s": 12.611169815063477}
+{"event": "train", "step": 180, "loss": 4.690134406089783, "learning_rate": 0.00023255813953488373, "elapsed_s": 13.094951391220093}
+{"event": "train", "step": 200, "loss": 4.681916737556458, "learning_rate": 0.00025839793281653745, "elapsed_s": 13.590264081954956}
+{"event": "train", "step": 220, "loss": 4.629971766471863, "learning_rate": 0.0002842377260981912, "elapsed_s": 14.047885179519653}
+{"event": "train", "step": 240, "loss": 4.655929899215698, "learning_rate": 0.000310077519379845, "elapsed_s": 14.495211601257324}
+{"event": "train", "step": 260, "loss": 4.6454051494598385, "learning_rate": 0.0003359173126614987, "elapsed_s": 14.940953969955444}
+{"event": "train", "step": 280, "loss": 4.675015878677368, "learning_rate": 0.00036175710594315246, "elapsed_s": 15.399949312210083}
+{"event": "train", "step": 300, "loss": 4.602758407592773, "learning_rate": 0.00038759689922480625, "elapsed_s": 15.874937057495117}
+{"event": "train", "step": 320, "loss": 4.603246212005615, "learning_rate": 0.00041343669250645994, "elapsed_s": 16.343674659729004}
+{"event": "train", "step": 340, "loss": 4.562363028526306, "learning_rate": 0.0004392764857881137, "elapsed_s": 16.81684923171997}
+{"event": "train", "step": 360, "loss": 4.67993643283844, "learning_rate": 0.00046511627906976747, "elapsed_s": 17.288190364837646}
+{"event": "train", "step": 380, "loss": 4.600531458854675, "learning_rate": 0.0004909560723514212, "elapsed_s": 17.75168013572693}
+{"event": "train", "step": 400, "loss": 4.563110518455505, "learning_rate": 0.000499998668401834, "elapsed_s": 18.2187922000885}
+{"event": "train", "step": 420, "loss": 4.555332779884338, "learning_rate": 0.0004999914195065388, "elapsed_s": 18.70127534866333}
+{"event": "train", "step": 440, "loss": 4.578880071640015, "learning_rate": 0.0004999778674119666, "elapsed_s": 19.177801847457886}
+{"event": "train", "step": 460, "loss": 4.541761088371277, "learning_rate": 0.0004999580124598156, "elapsed_s": 19.650096893310547}
+{"event": "train", "step": 480, "loss": 4.496497774124146, "learning_rate": 0.0004999318551507018, "elapsed_s": 20.11230993270874}
+{"event": "train", "step": 500, "loss": 4.448557496070862, "learning_rate": 0.0004998993961441468, "elapsed_s": 20.57578206062317}
+{"event": "train", "step": 520, "loss": 4.510877346992492, "learning_rate": 0.0004998606362585609, "elapsed_s": 21.204406261444092}
+{"event": "train", "step": 540, "loss": 4.4744020938873295, "learning_rate": 0.000499815576471223, "elapsed_s": 21.677183866500854}
+{"event": "train", "step": 560, "loss": 4.487486839294434, "learning_rate": 0.0004997642179182553, "elapsed_s": 22.141172647476196}
+{"event": "train", "step": 580, "loss": 4.401848483085632, "learning_rate": 0.0004997065618945947, "elapsed_s": 22.60059928894043}
+{"event": "train", "step": 600, "loss": 4.420126438140869, "learning_rate": 0.0004996426098539612, "elapsed_s": 23.06651258468628}
+{"event": "train", "step": 620, "loss": 4.438009738922119, "learning_rate": 0.0004995723634088196, "elapsed_s": 23.534086227416992}
+{"event": "train", "step": 640, "loss": 4.375300168991089, "learning_rate": 0.0004994958243303403, "elapsed_s": 24.006122589111328}
+{"event": "train", "step": 660, "loss": 4.403384900093078, "learning_rate": 0.0004994129945483537, "elapsed_s": 24.461066246032715}
+{"event": "train", "step": 680, "loss": 4.395544338226318, "learning_rate": 0.0004993238761513019, "elapsed_s": 24.903075218200684}
+{"event": "train", "step": 700, "loss": 4.395400977134704, "learning_rate": 0.0004992284713861862, "elapsed_s": 25.349731922149658}
+{"event": "train", "step": 720, "loss": 4.369986510276794, "learning_rate": 0.0004991267826585101, "elapsed_s": 25.79615354537964}
+{"event": "train", "step": 740, "loss": 4.386384916305542, "learning_rate": 0.0004990188125322189, "elapsed_s": 26.233235597610474}
+{"event": "train", "step": 760, "loss": 4.341637229919433, "learning_rate": 0.0004989045637296348, "elapsed_s": 26.66955804824829}
+{"event": "train", "step": 780, "loss": 4.288287782669068, "learning_rate": 0.0004987840391313886, "elapsed_s": 27.106184005737305}
+{"event": "train", "step": 800, "loss": 4.345422887802124, "learning_rate": 0.000498657241776347, "elapsed_s": 27.550791025161743}
+{"event": "train", "step": 820, "loss": 4.333413553237915, "learning_rate": 0.0004985241748615355, "elapsed_s": 27.989400386810303}
+{"event": "train", "step": 840, "loss": 4.385653495788574, "learning_rate": 0.0004983848417420585, "elapsed_s": 28.429297924041748}
+{"event": "train", "step": 860, "loss": 4.299694037437439, "learning_rate": 0.0004982392459310141, "elapsed_s": 28.869856357574463}
+{"event": "train", "step": 880, "loss": 4.299236297607422, "learning_rate": 0.0004980873910994061, "elapsed_s": 29.357131958007812}
+{"event": "train", "step": 900, "loss": 4.243064069747925, "learning_rate": 0.000497929281076051, "elapsed_s": 29.82161831855774}
+{"event": "train", "step": 920, "loss": 4.267381477355957, "learning_rate": 0.0004977649198474814, "elapsed_s": 30.29192280769348}
+{"event": "train", "step": 940, "loss": 4.3336035966873165, "learning_rate": 0.0004975943115578461, "elapsed_s": 30.74757194519043}
+{"event": "train", "step": 960, "loss": 4.227601885795593, "learning_rate": 0.0004974174605088046, "elapsed_s": 31.204246044158936}
+{"event": "train", "step": 980, "loss": 4.213042378425598, "learning_rate": 0.0004972343711594198, "elapsed_s": 31.654828310012817}
+{"event": "train", "step": 1000, "loss": 4.230743837356568, "learning_rate": 0.0004970450481260446, "elapsed_s": 32.1207549571991}
+{"event": "validation", "step": 1000, "loss": 4.278658509254456, "recall_at_1": 0.014}
+{"event": "train", "step": 1020, "loss": 4.256582713127136, "learning_rate": 0.0004968494961822062, "elapsed_s": 41.10897731781006}
+{"event": "train", "step": 1040, "loss": 4.2881251811981205, "learning_rate": 0.0004966477202584854, "elapsed_s": 41.54715394973755}
+{"event": "train", "step": 1060, "loss": 4.232862639427185, "learning_rate": 0.0004964397254423919, "elapsed_s": 41.98839783668518}
+{"event": "train", "step": 1080, "loss": 4.270459508895874, "learning_rate": 0.000496225516978237, "elapsed_s": 42.44052696228027}
+{"event": "train", "step": 1100, "loss": 4.253033185005188, "learning_rate": 0.0004960051002670007, "elapsed_s": 42.88282513618469}
+{"event": "train", "step": 1120, "loss": 4.260646104812622, "learning_rate": 0.0004957784808661955, "elapsed_s": 43.33740758895874}
+{"event": "train", "step": 1140, "loss": 4.248541378974915, "learning_rate": 0.0004955456644897265, "elapsed_s": 43.77753496170044}
+{"event": "train", "step": 1160, "loss": 4.223201107978821, "learning_rate": 0.0004953066570077471, "elapsed_s": 44.21784019470215}
+{"event": "train", "step": 1180, "loss": 4.2370490550994875, "learning_rate": 0.0004950614644465114, "elapsed_s": 44.66560769081116}
+{"event": "train", "step": 1200, "loss": 4.207837963104248, "learning_rate": 0.0004948100929882219, "elapsed_s": 45.105462312698364}
+{"event": "train", "step": 1220, "loss": 4.22065908908844, "learning_rate": 0.0004945525489708736, "elapsed_s": 45.54220128059387}
+{"event": "train", "step": 1240, "loss": 4.20661678314209, "learning_rate": 0.0004942888388880946, "elapsed_s": 45.97769212722778}
+{"event": "train", "step": 1260, "loss": 4.263231682777405, "learning_rate": 0.0004940189693889818, "elapsed_s": 46.42819571495056}
+{"event": "train", "step": 1280, "loss": 4.192655301094055, "learning_rate": 0.0004937429472779336, "elapsed_s": 46.8852424621582}
+{"event": "train", "step": 1300, "loss": 4.207595729827881, "learning_rate": 0.0004934607795144786, "elapsed_s": 47.33617663383484}
+{"event": "train", "step": 1320, "loss": 4.174138808250428, "learning_rate": 0.0004931724732130993, "elapsed_s": 47.78315258026123}
+{"event": "train", "step": 1340, "loss": 4.1492071628570555, "learning_rate": 0.0004928780356430537, "elapsed_s": 48.22921657562256}
+{"event": "train", "step": 1360, "loss": 4.134703755378723, "learning_rate": 0.0004925774742281911, "elapsed_s": 48.68402624130249}
+{"event": "train", "step": 1380, "loss": 4.173884570598602, "learning_rate": 0.0004922707965467657, "elapsed_s": 49.129775524139404}
+{"event": "train", "step": 1400, "loss": 4.162477385997772, "learning_rate": 0.0004919580103312449, "elapsed_s": 49.58668828010559}
+{"event": "train", "step": 1420, "loss": 4.172889018058777, "learning_rate": 0.000491639123468115, "elapsed_s": 50.02582001686096}
+{"event": "train", "step": 1440, "loss": 4.200236535072326, "learning_rate": 0.0004913141439976816, "elapsed_s": 50.46493124961853}
+{"event": "train", "step": 1460, "loss": 4.130017423629761, "learning_rate": 0.0004909830801138673, "elapsed_s": 50.90474057197571}
+{"event": "train", "step": 1480, "loss": 4.14314444065094, "learning_rate": 0.0004906459401640053, "elapsed_s": 51.3602192401886}
+{"event": "train", "step": 1500, "loss": 4.12908376455307, "learning_rate": 0.0004903027326486286, "elapsed_s": 51.80715012550354}
+{"event": "train", "step": 1520, "loss": 4.14359986782074, "learning_rate": 0.0004899534662212553, "elapsed_s": 52.26277542114258}
+{"event": "train", "step": 1540, "loss": 4.134027004241943, "learning_rate": 0.0004895981496881715, "elapsed_s": 52.708080768585205}
+{"event": "train", "step": 1560, "loss": 4.116017496585846, "learning_rate": 0.0004892367920082082, "elapsed_s": 53.151955127716064}
+{"event": "train", "step": 1580, "loss": 4.153079652786255, "learning_rate": 0.0004888694022925161, "elapsed_s": 53.59013342857361}
+{"event": "train", "step": 1600, "loss": 4.116498851776123, "learning_rate": 0.0004884959898043354, "elapsed_s": 54.033042669296265}
+{"event": "train", "step": 1620, "loss": 4.135101675987244, "learning_rate": 0.0004881165639587627, "elapsed_s": 54.46528506278992}
+{"event": "train", "step": 1640, "loss": 4.185906529426575, "learning_rate": 0.00048773113432251295, "elapsed_s": 54.9075129032135}
+{"event": "train", "step": 1660, "loss": 4.159037971496582, "learning_rate": 0.0004873397106136791, "elapsed_s": 55.34476375579834}
+{"event": "train", "step": 1680, "loss": 4.12104344367981, "learning_rate": 0.0004869423027014864, "elapsed_s": 55.783405780792236}
+{"event": "train", "step": 1700, "loss": 4.100151073932648, "learning_rate": 0.0004865389206060438, "elapsed_s": 56.22089719772339}
+{"event": "train", "step": 1720, "loss": 4.147478127479554, "learning_rate": 0.00048612957449809137, "elapsed_s": 56.65522623062134}
+{"event": "train", "step": 1740, "loss": 4.09362610578537, "learning_rate": 0.00048571427469874356, "elapsed_s": 57.11369824409485}
+{"event": "train", "step": 1760, "loss": 4.103320121765137, "learning_rate": 0.0004852930316792296, "elapsed_s": 57.56131434440613}
+{"event": "train", "step": 1780, "loss": 4.107409548759461, "learning_rate": 0.0004848658560606285, "elapsed_s": 58.006327390670776}
+{"event": "train", "step": 1800, "loss": 4.100596714019775, "learning_rate": 0.00048443275861360236, "elapsed_s": 58.448675870895386}
+{"event": "train", "step": 1820, "loss": 4.125512647628784, "learning_rate": 0.0004839937502581239, "elapsed_s": 58.90279984474182}
+{"event": "train", "step": 1840, "loss": 4.110939335823059, "learning_rate": 0.00048354884206320176, "elapsed_s": 59.35634136199951}
+{"event": "train", "step": 1860, "loss": 4.106650257110596, "learning_rate": 0.00048309804524660096, "elapsed_s": 59.80231690406799}
+{"event": "train", "step": 1880, "loss": 4.073759281635285, "learning_rate": 0.0004826413711745602, "elapsed_s": 60.244447469711304}
+{"event": "train", "step": 1900, "loss": 4.105740714073181, "learning_rate": 0.00048217883136150545, "elapsed_s": 60.68072700500488}
+{"event": "train", "step": 1920, "loss": 4.082316827774048, "learning_rate": 0.0004817104374697595, "elapsed_s": 61.13827967643738}
+{"event": "train", "step": 1940, "loss": 4.090257573127746, "learning_rate": 0.00048123620130924777, "elapsed_s": 61.578453063964844}
+{"event": "train", "step": 1960, "loss": 4.082735657691956, "learning_rate": 0.0004807561348372008, "elapsed_s": 62.023608922958374}
+{"event": "train", "step": 1980, "loss": 4.089769732952118, "learning_rate": 0.0004802702501578526, "elapsed_s": 62.46874165534973}
+{"event": "train", "step": 2000, "loss": 4.042151069641113, "learning_rate": 0.0004797785595221354, "elapsed_s": 62.918880462646484}
+{"event": "validation", "step": 2000, "loss": 4.103663325309753, "recall_at_1": 0.008}
+{"event": "train", "step": 2020, "loss": 4.04916878938675, "learning_rate": 0.0004792810753273711, "elapsed_s": 63.611833333969116}
+{"event": "train", "step": 2040, "loss": 4.084432446956635, "learning_rate": 0.0004787778101169581, "elapsed_s": 64.05853271484375}
+{"event": "train", "step": 2060, "loss": 4.080963397026062, "learning_rate": 0.00047826877658005564, "elapsed_s": 64.50654292106628}
+{"event": "train", "step": 2080, "loss": 4.0441605806350704, "learning_rate": 0.0004777539875512635, "elapsed_s": 64.9576575756073}
+{"event": "train", "step": 2100, "loss": 4.064308547973633, "learning_rate": 0.00047723345601029856, "elapsed_s": 65.39872527122498}
+{"event": "train", "step": 2120, "loss": 4.070480215549469, "learning_rate": 0.0004767071950816673, "elapsed_s": 65.84652519226074}
+{"event": "train", "step": 2140, "loss": 4.05004643201828, "learning_rate": 0.0004761752180343353, "elapsed_s": 66.28458452224731}
+{"event": "train", "step": 2160, "loss": 4.039112281799317, "learning_rate": 0.0004756375382813921, "elapsed_s": 66.74167776107788}
+{"event": "train", "step": 2180, "loss": 4.058060801029205, "learning_rate": 0.0004750941693797137, "elapsed_s": 67.31537890434265}
+{"event": "train", "step": 2200, "loss": 4.044344687461853, "learning_rate": 0.0004745451250296201, "elapsed_s": 67.77072787284851}
+{"event": "train", "step": 2220, "loss": 4.022262680530548, "learning_rate": 0.0004739904190745303, "elapsed_s": 68.21583104133606}
+{"event": "train", "step": 2240, "loss": 3.9987276077270506, "learning_rate": 0.00047343006550061296, "elapsed_s": 68.67972826957703}
+{"event": "train", "step": 2260, "loss": 4.033700263500213, "learning_rate": 0.000472864078436434, "elapsed_s": 69.12975144386292}
+{"event": "train", "step": 2280, "loss": 4.010021650791169, "learning_rate": 0.0004722924721526002, "elapsed_s": 69.58771133422852}
+{"event": "train", "step": 2300, "loss": 4.02365140914917, "learning_rate": 0.0004717152610613995, "elapsed_s": 70.03674221038818}
+{"event": "train", "step": 2320, "loss": 4.017482101917267, "learning_rate": 0.0004711324597164376, "elapsed_s": 70.4855444431305}
+{"event": "train", "step": 2340, "loss": 4.009071695804596, "learning_rate": 0.0004705440828122708, "elapsed_s": 70.93550753593445}
+{"event": "train", "step": 2360, "loss": 4.018468487262726, "learning_rate": 0.00046995014518403576, "elapsed_s": 71.38122534751892}
+{"event": "train", "step": 2380, "loss": 3.991307282447815, "learning_rate": 0.00046935066180707556, "elapsed_s": 71.81735348701477}
+{"event": "train", "step": 2400, "loss": 4.044314205646515, "learning_rate": 0.0004687456477965616, "elapsed_s": 72.25459456443787}
+{"event": "train", "step": 2420, "loss": 4.032532012462616, "learning_rate": 0.000468135118407113, "elapsed_s": 72.69027876853943}
+{"event": "train", "step": 2440, "loss": 4.044214916229248, "learning_rate": 0.0004675190890324116, "elapsed_s": 73.13457226753235}
+{"event": "train", "step": 2460, "loss": 3.9999911189079285, "learning_rate": 0.00046689757520481425, "elapsed_s": 73.5788824558258}
+{"event": "train", "step": 2480, "loss": 3.979391634464264, "learning_rate": 0.00046627059259496086, "elapsed_s": 74.02644419670105}
+{"event": "train", "step": 2500, "loss": 4.021868813037872, "learning_rate": 0.00046563815701137926, "elapsed_s": 74.46334195137024}
+{"event": "train", "step": 2520, "loss": 3.9982270121574404, "learning_rate": 0.0004650002844000868, "elapsed_s": 74.89837145805359}
+{"event": "train", "step": 2540, "loss": 4.022631895542145, "learning_rate": 0.00046435699084418835, "elapsed_s": 75.3425087928772}
+{"event": "train", "step": 2560, "loss": 4.038698315620422, "learning_rate": 0.00046370829256347037, "elapsed_s": 75.7784583568573}
+{"event": "train", "step": 2580, "loss": 4.029807806015015, "learning_rate": 0.0004630542059139923, "elapsed_s": 76.21102476119995}
+{"event": "train", "step": 2600, "loss": 3.964472734928131, "learning_rate": 0.0004623947473876744, "elapsed_s": 76.66975402832031}
+{"event": "train", "step": 2620, "loss": 3.964149272441864, "learning_rate": 0.00046172993361188113, "elapsed_s": 77.10862231254578}
+{"event": "train", "step": 2640, "loss": 3.9854819059371946, "learning_rate": 0.00046105978134900285, "elapsed_s": 77.54387879371643}
+{"event": "train", "step": 2660, "loss": 3.976743149757385, "learning_rate": 0.0004603843074960324, "elapsed_s": 77.98907732963562}
+{"event": "train", "step": 2680, "loss": 3.9553184151649474, "learning_rate": 0.0004597035290841396, "elapsed_s": 78.43977546691895}
+{"event": "train", "step": 2700, "loss": 3.9782107949256895, "learning_rate": 0.00045901746327824155, "elapsed_s": 78.87921571731567}
+{"event": "train", "step": 2720, "loss": 3.9343753457069397, "learning_rate": 0.00045832612737656985, "elapsed_s": 79.31299686431885}
+{"event": "train", "step": 2740, "loss": 3.982185685634613, "learning_rate": 0.00045762953881023474, "elapsed_s": 79.74914288520813}
+{"event": "train", "step": 2760, "loss": 3.9614883780479433, "learning_rate": 0.0004569277151427851, "elapsed_s": 80.20457172393799}
+{"event": "train", "step": 2780, "loss": 3.9674476742744447, "learning_rate": 0.000456220674069766, "elapsed_s": 80.64596891403198}
+{"event": "train", "step": 2800, "loss": 3.963286566734314, "learning_rate": 0.00045550843341827255, "elapsed_s": 81.0956482887268}
+{"event": "train", "step": 2820, "loss": 3.9954036235809327, "learning_rate": 0.0004547910111465, "elapsed_s": 81.54567289352417}
+{"event": "train", "step": 2840, "loss": 3.9570708870887756, "learning_rate": 0.0004540684253432915, "elapsed_s": 81.98324131965637}
+{"event": "train", "step": 2860, "loss": 3.986406624317169, "learning_rate": 0.00045334069422768147, "elapsed_s": 82.42099404335022}
+{"event": "train", "step": 2880, "loss": 3.928980839252472, "learning_rate": 0.00045260783614843684, "elapsed_s": 82.85836553573608}
+{"event": "train", "step": 2900, "loss": 3.969514036178589, "learning_rate": 0.0004518698695835936, "elapsed_s": 83.30556297302246}
+{"event": "train", "step": 2920, "loss": 3.9822160720825197, "learning_rate": 0.0004511268131399919, "elapsed_s": 83.77758049964905}
+{"event": "train", "step": 2940, "loss": 3.9744591116905212, "learning_rate": 0.000450378685552806, "elapsed_s": 84.21833443641663}
+{"event": "train", "step": 2960, "loss": 3.916402530670166, "learning_rate": 0.0004496255056850724, "elapsed_s": 84.65928673744202}
+{"event": "train", "step": 2980, "loss": 3.980082070827484, "learning_rate": 0.00044886729252721413, "elapsed_s": 85.10070133209229}
+{"event": "train", "step": 3000, "loss": 3.9875089645385744, "learning_rate": 0.00044810406519656175, "elapsed_s": 85.55433773994446}
+{"event": "validation", "step": 3000, "loss": 4.053720921278, "recall_at_1": 0.0105}
+{"event": "train", "step": 3020, "loss": 4.005371165275574, "learning_rate": 0.0004473358429368717, "elapsed_s": 86.30154824256897}
+{"event": "train", "step": 3040, "loss": 3.913334858417511, "learning_rate": 0.0004465626451178406, "elapsed_s": 86.74464869499207}
+{"event": "train", "step": 3060, "loss": 3.9280787229537966, "learning_rate": 0.0004457844912346174, "elapsed_s": 87.18313217163086}
+{"event": "train", "step": 3080, "loss": 3.9135559439659118, "learning_rate": 0.00044500140090731146, "elapsed_s": 87.62471294403076}
+{"event": "train", "step": 3100, "loss": 3.9331393599510194, "learning_rate": 0.00044421339388049804, "elapsed_s": 88.080162525177}
+{"event": "train", "step": 3120, "loss": 3.936568021774292, "learning_rate": 0.00044342049002272024, "elapsed_s": 88.52393817901611}
+{"event": "train", "step": 3140, "loss": 3.9659378051757814, "learning_rate": 0.00044262270932598835, "elapsed_s": 88.99354696273804}
+{"event": "train", "step": 3160, "loss": 3.9631355881690977, "learning_rate": 0.00044182007190527573, "elapsed_s": 89.44700622558594}
+{"event": "train", "step": 3180, "loss": 3.908782410621643, "learning_rate": 0.00044101259799801126, "elapsed_s": 89.89410066604614}
+{"event": "train", "step": 3200, "loss": 3.877232003211975, "learning_rate": 0.0004402003079635695, "elapsed_s": 90.34263443946838}
+{"event": "train", "step": 3220, "loss": 3.918678569793701, "learning_rate": 0.0004393832222827573, "elapsed_s": 90.8019700050354}
+{"event": "train", "step": 3240, "loss": 3.904724133014679, "learning_rate": 0.00043856136155729725, "elapsed_s": 91.25120449066162}
+{"event": "train", "step": 3260, "loss": 3.8961193442344664, "learning_rate": 0.0004377347465093082, "elapsed_s": 91.7101047039032}
+{"event": "train", "step": 3280, "loss": 3.9407534837722777, "learning_rate": 0.00043690339798078314, "elapsed_s": 92.15341448783875}
+{"event": "train", "step": 3300, "loss": 3.955256199836731, "learning_rate": 0.0004360673369330631, "elapsed_s": 92.59773540496826}
+{"event": "train", "step": 3320, "loss": 3.924412655830383, "learning_rate": 0.00043522658444630926, "elapsed_s": 93.06260824203491}
+{"event": "train", "step": 3340, "loss": 3.9289205431938172, "learning_rate": 0.000434381161718971, "elapsed_s": 93.52145671844482}
+{"event": "train", "step": 3360, "loss": 3.9186713814735414, "learning_rate": 0.00043353109006725175, "elapsed_s": 93.98447632789612}
+{"event": "train", "step": 3380, "loss": 3.9647103190422057, "learning_rate": 0.0004326763909245711, "elapsed_s": 94.43496704101562}
+{"event": "train", "step": 3400, "loss": 3.894024705886841, "learning_rate": 0.00043181708584102475, "elapsed_s": 94.88772559165955}
+{"event": "train", "step": 3420, "loss": 3.936267948150635, "learning_rate": 0.0004309531964828412, "elapsed_s": 95.34637093544006}
+{"event": "train", "step": 3440, "loss": 3.923455798625946, "learning_rate": 0.000430084744631835, "elapsed_s": 95.7895405292511}
+{"event": "train", "step": 3460, "loss": 3.881089186668396, "learning_rate": 0.00042921175218485816, "elapsed_s": 96.27558827400208}
+{"event": "train", "step": 3480, "loss": 3.8935436487197874, "learning_rate": 0.00042833424115324743, "elapsed_s": 96.72868418693542}
+{"event": "train", "step": 3500, "loss": 3.8669028878211975, "learning_rate": 0.00042745223366226986, "elapsed_s": 97.18549656867981}
+{"event": "train", "step": 3520, "loss": 3.8806556344032286, "learning_rate": 0.0004265657519505647, "elapsed_s": 97.64035320281982}
+{"event": "train", "step": 3540, "loss": 3.909578025341034, "learning_rate": 0.0004256748183695826, "elapsed_s": 98.08388757705688}
+{"event": "train", "step": 3560, "loss": 3.919867420196533, "learning_rate": 0.0004247794553830221, "elapsed_s": 98.54107451438904}
+{"event": "train", "step": 3580, "loss": 3.884328877925873, "learning_rate": 0.0004238796855662634, "elapsed_s": 99.0127363204956}
+{"event": "train", "step": 3600, "loss": 3.8811463832855226, "learning_rate": 0.00042297553160579906, "elapsed_s": 99.46459341049194}
+{"event": "train", "step": 3620, "loss": 3.8923567175865172, "learning_rate": 0.00042206701629866173, "elapsed_s": 99.90228867530823}
+{"event": "train", "step": 3640, "loss": 3.8866697549819946, "learning_rate": 0.0004211541625518499, "elapsed_s": 100.34108304977417}
+{"event": "train", "step": 3660, "loss": 3.9368638038635253, "learning_rate": 0.0004202369933817497, "elapsed_s": 100.79042720794678}
+{"event": "train", "step": 3680, "loss": 3.868456983566284, "learning_rate": 0.0004193155319135551, "elapsed_s": 101.23855924606323}
+{"event": "train", "step": 3700, "loss": 3.8643144607543944, "learning_rate": 0.00041838980138068444, "elapsed_s": 101.67513275146484}
+{"event": "train", "step": 3720, "loss": 3.8935752511024475, "learning_rate": 0.0004174598251241951, "elapsed_s": 102.12298154830933}
+{"event": "train", "step": 3740, "loss": 3.887183463573456, "learning_rate": 0.00041652562659219447, "elapsed_s": 102.56008076667786}
+{"event": "train", "step": 3760, "loss": 3.8448415279388426, "learning_rate": 0.00041558722933924896, "elapsed_s": 103.02388548851013}
+{"event": "train", "step": 3780, "loss": 3.868120884895325, "learning_rate": 0.00041464465702579036, "elapsed_s": 103.46505808830261}
+{"event": "train", "step": 3800, "loss": 3.893366885185242, "learning_rate": 0.00041369793341751866, "elapsed_s": 103.91359066963196}
+{"event": "train", "step": 3820, "loss": 3.906751847267151, "learning_rate": 0.0004127470823848035, "elapsed_s": 104.35450959205627}
+{"event": "train", "step": 3840, "loss": 3.869689738750458, "learning_rate": 0.0004117921279020819, "elapsed_s": 104.80509209632874}
+{"event": "train", "step": 3860, "loss": 3.887068819999695, "learning_rate": 0.00041083309404725386, "elapsed_s": 105.24388670921326}
+{"event": "train", "step": 3880, "loss": 3.8691829562187197, "learning_rate": 0.0004098700050010753, "elapsed_s": 105.814523935318}
+{"event": "train", "step": 3900, "loss": 3.7998663544654847, "learning_rate": 0.00040890288504654836, "elapsed_s": 106.27005743980408}
+{"event": "train", "step": 3920, "loss": 3.8066797852516174, "learning_rate": 0.00040793175856830924, "elapsed_s": 106.72138524055481}
+{"event": "train", "step": 3940, "loss": 3.8411558151245115, "learning_rate": 0.0004069566500520131, "elapsed_s": 107.16075229644775}
+{"event": "train", "step": 3960, "loss": 3.8191500663757325, "learning_rate": 0.00040597758408371707, "elapsed_s": 107.60193395614624}
+{"event": "train", "step": 3980, "loss": 3.8186970472335817, "learning_rate": 0.0004049945853492599, "elapsed_s": 108.0532112121582}
+{"event": "train", "step": 4000, "loss": 3.827391970157623, "learning_rate": 0.00040400767863363996, "elapsed_s": 108.51269555091858}
+{"event": "validation", "step": 4000, "loss": 4.107629656791687, "recall_at_1": 0.015}
+{"event": "train", "step": 4020, "loss": 3.8832298040390016, "learning_rate": 0.00040301688882039014, "elapsed_s": 109.28891944885254}
+{"event": "train", "step": 4040, "loss": 3.830469381809235, "learning_rate": 0.0004020222408909503, "elapsed_s": 109.73159384727478}
+{"event": "train", "step": 4060, "loss": 3.832515871524811, "learning_rate": 0.00040102375992403773, "elapsed_s": 110.18351602554321}
+{"event": "train", "step": 4080, "loss": 3.8747861623764037, "learning_rate": 0.0004000214710950145, "elapsed_s": 110.63763952255249}
+{"event": "train", "step": 4100, "loss": 3.816035747528076, "learning_rate": 0.0003990153996752528, "elapsed_s": 111.08085823059082}
+{"event": "train", "step": 4120, "loss": 3.8279685735702516, "learning_rate": 0.0003980055710314979, "elapsed_s": 111.53152585029602}
+{"event": "train", "step": 4140, "loss": 3.8220778465270997, "learning_rate": 0.00039699201062522844, "elapsed_s": 111.99048042297363}
+{"event": "train", "step": 4160, "loss": 3.8508506774902345, "learning_rate": 0.00039597474401201427, "elapsed_s": 112.43756365776062}
+{"event": "train", "step": 4180, "loss": 3.838638651371002, "learning_rate": 0.0003949537968408726, "elapsed_s": 112.87789273262024}
+{"event": "train", "step": 4200, "loss": 3.847361218929291, "learning_rate": 0.00039392919485362076, "elapsed_s": 113.31759428977966}
+{"event": "train", "step": 4220, "loss": 3.8406118869781496, "learning_rate": 0.00039290096388422743, "elapsed_s": 113.76220750808716}
+{"event": "train", "step": 4240, "loss": 3.8668089747428893, "learning_rate": 0.00039186912985816126, "elapsed_s": 114.21372509002686}
+{"event": "train", "step": 4260, "loss": 3.8429593563079836, "learning_rate": 0.000390833718791737, "elapsed_s": 114.65712475776672}
+{"event": "train", "step": 4280, "loss": 3.8897117376327515, "learning_rate": 0.00038979475679146007, "elapsed_s": 115.10719537734985}
+{"event": "train", "step": 4300, "loss": 3.846380126476288, "learning_rate": 0.0003887522700533675, "elapsed_s": 115.54892253875732}
+{"event": "train", "step": 4320, "loss": 3.7951263666152952, "learning_rate": 0.0003877062848623681, "elapsed_s": 116.00970101356506}
+{"event": "train", "step": 4340, "loss": 3.7864400148391724, "learning_rate": 0.00038665682759157945, "elapsed_s": 116.44790530204773}
+{"event": "train", "step": 4360, "loss": 3.7935109376907348, "learning_rate": 0.00038560392470166294, "elapsed_s": 116.88503766059875}
+{"event": "train", "step": 4380, "loss": 3.815704679489136, "learning_rate": 0.0003845476027401568, "elapsed_s": 117.3260931968689}
+{"event": "train", "step": 4400, "loss": 3.7780648708343505, "learning_rate": 0.00038348788834080614, "elapsed_s": 117.78047370910645}
+{"event": "train", "step": 4420, "loss": 3.795751988887787, "learning_rate": 0.0003824248082228925, "elapsed_s": 118.23186135292053}
+{"event": "train", "step": 4440, "loss": 3.7649246573448183, "learning_rate": 0.000381358389190559, "elapsed_s": 118.69606065750122}
+{"event": "train", "step": 4460, "loss": 3.7649667739868162, "learning_rate": 0.0003802886581321353, "elapsed_s": 119.13991475105286}
+{"event": "train", "step": 4480, "loss": 3.799170672893524, "learning_rate": 0.0003792156420194591, "elapsed_s": 119.58278703689575}
+{"event": "train", "step": 4500, "loss": 3.8014742970466613, "learning_rate": 0.0003781393679071967, "elapsed_s": 120.02610468864441}
+{"event": "train", "step": 4520, "loss": 3.826338756084442, "learning_rate": 0.0003770598629321602, "elapsed_s": 120.49128127098083}
+{"event": "train", "step": 4540, "loss": 3.759580969810486, "learning_rate": 0.0003759771543126237, "elapsed_s": 120.96463823318481}
+{"event": "train", "step": 4560, "loss": 3.8210657119750975, "learning_rate": 0.00037489126934763694, "elapsed_s": 121.47085118293762}
+{"event": "train", "step": 4580, "loss": 3.8196467995643615, "learning_rate": 0.0003738022354163369, "elapsed_s": 121.94458532333374}
+{"event": "train", "step": 4600, "loss": 3.821422517299652, "learning_rate": 0.00037271007997725756, "elapsed_s": 122.39595890045166}
+{"event": "train", "step": 4620, "loss": 3.814608335494995, "learning_rate": 0.0003716148305676376, "elapsed_s": 122.85207748413086}
+{"event": "train", "step": 4640, "loss": 3.789643633365631, "learning_rate": 0.00037051651480272594, "elapsed_s": 123.29517602920532}
+{"event": "train", "step": 4660, "loss": 3.805340027809143, "learning_rate": 0.0003694151603750856, "elapsed_s": 123.74942970275879}
+{"event": "train", "step": 4680, "loss": 3.7968928933143617, "learning_rate": 0.0003683107950538954, "elapsed_s": 124.2225923538208}
+{"event": "train", "step": 4700, "loss": 3.7704756379127504, "learning_rate": 0.0003672034466842498, "elapsed_s": 124.71569800376892}
+{"event": "train", "step": 4720, "loss": 3.7683708906173705, "learning_rate": 0.0003660931431864567, "elapsed_s": 125.16684198379517}
+{"event": "train", "step": 4740, "loss": 3.7737975120544434, "learning_rate": 0.000364979912555334, "elapsed_s": 125.61912250518799}
+{"event": "train", "step": 4760, "loss": 3.6990949273109437, "learning_rate": 0.0003638637828595031, "elapsed_s": 126.06542015075684}
+{"event": "train", "step": 4780, "loss": 3.754763424396515, "learning_rate": 0.0003627447822406813, "elapsed_s": 126.51020812988281}
+{"event": "train", "step": 4800, "loss": 3.7273942112922667, "learning_rate": 0.00036162293891297257, "elapsed_s": 126.98047828674316}
+{"event": "train", "step": 4820, "loss": 3.7335302114486693, "learning_rate": 0.000360498281162156, "elapsed_s": 127.44014859199524}
+{"event": "train", "step": 4840, "loss": 3.8014326810836794, "learning_rate": 0.00035937083734497246, "elapsed_s": 127.90865898132324}
+{"event": "train", "step": 4860, "loss": 3.7420841336250303, "learning_rate": 0.0003582406358884098, "elapsed_s": 128.36916184425354}
+{"event": "train", "step": 4880, "loss": 3.7228352665901183, "learning_rate": 0.0003571077052889863, "elapsed_s": 128.83107328414917}
+{"event": "train", "step": 4900, "loss": 3.743587303161621, "learning_rate": 0.00035597207411203153, "elapsed_s": 129.27693510055542}
+{"event": "train", "step": 4920, "loss": 3.765555369853973, "learning_rate": 0.0003548337709909667, "elapsed_s": 129.74150323867798}
+{"event": "train", "step": 4940, "loss": 3.759941852092743, "learning_rate": 0.0003536928246265826, "elapsed_s": 130.19663166999817}
+{"event": "train", "step": 4960, "loss": 3.731901836395264, "learning_rate": 0.00035254926378631567, "elapsed_s": 130.65585780143738}
+{"event": "train", "step": 4980, "loss": 3.772131395339966, "learning_rate": 0.00035140311730352315, "elapsed_s": 131.11717557907104}
+{"event": "train", "step": 5000, "loss": 3.750517416000366, "learning_rate": 0.0003502544140767555, "elapsed_s": 131.57847452163696}
+{"event": "validation", "step": 5000, "loss": 4.0789260268211365, "recall_at_1": 0.0125}
+{"event": "train", "step": 5020, "loss": 3.7867310285568236, "learning_rate": 0.0003491031830690282, "elapsed_s": 132.39796423912048}
+{"event": "train", "step": 5040, "loss": 3.7300582408905028, "learning_rate": 0.00034794945330709126, "elapsed_s": 132.85017704963684}
+{"event": "train", "step": 5060, "loss": 3.7720590829849243, "learning_rate": 0.0003467932538806976, "elapsed_s": 133.3143081665039}
+{"event": "train", "step": 5080, "loss": 3.763454520702362, "learning_rate": 0.00034563461394186934, "elapsed_s": 133.7845494747162}
+{"event": "train", "step": 5100, "loss": 3.7772231340408324, "learning_rate": 0.0003444735627041627, "elapsed_s": 134.24632096290588}
+{"event": "train", "step": 5120, "loss": 3.758765661716461, "learning_rate": 0.0003433101294419319, "elapsed_s": 134.69726729393005}
+{"event": "train", "step": 5140, "loss": 3.7626577019691467, "learning_rate": 0.00034214434348959046, "elapsed_s": 135.1556453704834}
+{"event": "train", "step": 5160, "loss": 3.7429693937301636, "learning_rate": 0.0003409762342408719, "elapsed_s": 135.58728909492493}
+{"event": "train", "step": 5180, "loss": 3.6851349472999573, "learning_rate": 0.00033980583114808855, "elapsed_s": 136.05952835083008}
+{"event": "train", "step": 5200, "loss": 3.678535985946655, "learning_rate": 0.00033863316372138906, "elapsed_s": 136.50041222572327}
+{"event": "train", "step": 5220, "loss": 3.6733216762542726, "learning_rate": 0.0003374582615280141, "elapsed_s": 136.94266366958618}
+{"event": "train", "step": 5240, "loss": 3.6608057260513305, "learning_rate": 0.000336281154191551, "elapsed_s": 137.3938479423523}
+{"event": "train", "step": 5260, "loss": 3.690261113643646, "learning_rate": 0.0003351018713911869, "elapsed_s": 137.85861921310425}
+{"event": "train", "step": 5280, "loss": 3.7006960988044737, "learning_rate": 0.0003339204428609602, "elapsed_s": 138.30762195587158}
+{"event": "train", "step": 5300, "loss": 3.6953903079032897, "learning_rate": 0.00033273689838901133, "elapsed_s": 138.76750326156616}
+{"event": "train", "step": 5320, "loss": 3.7487650513648987, "learning_rate": 0.0003315512678168309, "elapsed_s": 139.21841430664062}
+{"event": "train", "step": 5340, "loss": 3.668799042701721, "learning_rate": 0.0003303635810385082, "elapsed_s": 139.6747488975525}
+{"event": "train", "step": 5360, "loss": 3.7033811688423155, "learning_rate": 0.00032917386799997673, "elapsed_s": 140.13376665115356}
+{"event": "train", "step": 5380, "loss": 3.676067221164703, "learning_rate": 0.0003279821586982594, "elapsed_s": 140.5984241962433}
+{"event": "train", "step": 5400, "loss": 3.7175679922103884, "learning_rate": 0.00032678848318071246, "elapsed_s": 141.0529227256775}
+{"event": "train", "step": 5420, "loss": 3.658122682571411, "learning_rate": 0.0003255928715442674, "elapsed_s": 141.54012751579285}
+{"event": "train", "step": 5440, "loss": 3.763239824771881, "learning_rate": 0.0003243953539346728, "elapsed_s": 142.03496265411377}
+{"event": "train", "step": 5460, "loss": 3.7305675745010376, "learning_rate": 0.0003231959605457331, "elapsed_s": 142.53136467933655}
+{"event": "train", "step": 5480, "loss": 3.692132353782654, "learning_rate": 0.00032199472161854864, "elapsed_s": 143.04246473312378}
+{"event": "train", "step": 5500, "loss": 3.7282273054122923, "learning_rate": 0.00032079166744075215, "elapsed_s": 143.5061433315277}
+{"event": "train", "step": 5520, "loss": 3.68942734003067, "learning_rate": 0.0003195868283457458, "elapsed_s": 143.9645459651947}
+{"event": "train", "step": 5540, "loss": 3.7403520584106444, "learning_rate": 0.0003183802347119358, "elapsed_s": 144.43579983711243}
+{"event": "train", "step": 5560, "loss": 3.7081547021865844, "learning_rate": 0.0003171719169619668, "elapsed_s": 144.88634610176086}
+{"event": "train", "step": 5580, "loss": 3.7678179264068605, "learning_rate": 0.0003159619055619547, "elapsed_s": 145.47090435028076}
+{"event": "train", "step": 5600, "loss": 3.718860757350922, "learning_rate": 0.0003147502310207186, "elapsed_s": 145.94973516464233}
+{"event": "train", "step": 5620, "loss": 3.6176072359085083, "learning_rate": 0.0003135369238890114, "elapsed_s": 146.41084122657776}
+{"event": "train", "step": 5640, "loss": 3.6284944295883177, "learning_rate": 0.00031232201475874953, "elapsed_s": 146.8528335094452}
+{"event": "train", "step": 5660, "loss": 3.6672277569770815, "learning_rate": 0.0003111055342622419, "elapsed_s": 147.29595375061035}
+{"event": "train", "step": 5680, "loss": 3.620872211456299, "learning_rate": 0.0003098875130714172, "elapsed_s": 147.75511384010315}
+{"event": "train", "step": 5700, "loss": 3.639104974269867, "learning_rate": 0.0003086679818970504, "elapsed_s": 148.21391773223877}
+{"event": "train", "step": 5720, "loss": 3.6934951186180114, "learning_rate": 0.0003074469714879892, "elapsed_s": 148.6592242717743}
+{"event": "train", "step": 5740, "loss": 3.672604191303253, "learning_rate": 0.00030622451263037783, "elapsed_s": 149.10393929481506}
+{"event": "train", "step": 5760, "loss": 3.6640511631965635, "learning_rate": 0.00030500063614688136, "elapsed_s": 149.57072734832764}
+{"event": "train", "step": 5780, "loss": 3.6843728542327883, "learning_rate": 0.0003037753728959082, "elapsed_s": 150.0325632095337}
+{"event": "train", "step": 5800, "loss": 3.646111619472504, "learning_rate": 0.0003025487537708327, "elapsed_s": 150.482825756073}
+{"event": "train", "step": 5820, "loss": 3.636573076248169, "learning_rate": 0.0003013208096992154, "elapsed_s": 150.93290638923645}
+{"event": "train", "step": 5840, "loss": 3.6668641686439516, "learning_rate": 0.0003000915716420235, "elapsed_s": 151.38690161705017}
+{"event": "train", "step": 5860, "loss": 3.6866225481033323, "learning_rate": 0.0002988610705928507, "elapsed_s": 151.83494901657104}
+{"event": "train", "step": 5880, "loss": 3.6608437180519102, "learning_rate": 0.0002976293375771351, "elapsed_s": 152.3029854297638}
+{"event": "train", "step": 5900, "loss": 3.7051465272903443, "learning_rate": 0.00029639640365137723, "elapsed_s": 152.76299738883972}
+{"event": "train", "step": 5920, "loss": 3.6644269824028015, "learning_rate": 0.00029516229990235687, "elapsed_s": 153.20716953277588}
+{"event": "train", "step": 5940, "loss": 3.657090401649475, "learning_rate": 0.0002939270574463495, "elapsed_s": 153.66088247299194}
+{"event": "train", "step": 5960, "loss": 3.6520742416381835, "learning_rate": 0.00029269070742834156, "elapsed_s": 154.1217794418335}
+{"event": "train", "step": 5980, "loss": 3.681882607936859, "learning_rate": 0.0002914532810212451, "elapsed_s": 154.5893154144287}
+{"event": "train", "step": 6000, "loss": 3.6749319195747376, "learning_rate": 0.00029021480942511185, "elapsed_s": 155.03552341461182}
+{"event": "validation", "step": 6000, "loss": 4.011619985103607, "recall_at_1": 0.013}
+{"event": "train", "step": 6020, "loss": 3.677917516231537, "learning_rate": 0.00028897532386634663, "elapsed_s": 155.71769332885742}
+{"event": "train", "step": 6040, "loss": 3.5605270147323607, "learning_rate": 0.00028773485559692, "elapsed_s": 156.21532034873962}
+{"event": "train", "step": 6060, "loss": 3.6248206615448, "learning_rate": 0.0002864934358935803, "elapsed_s": 156.66144847869873}
+{"event": "train", "step": 6080, "loss": 3.5675201296806334, "learning_rate": 0.00028525109605706485, "elapsed_s": 157.14078998565674}
+{"event": "train", "step": 6100, "loss": 3.599752902984619, "learning_rate": 0.00028400786741131104, "elapsed_s": 157.63578367233276}
+{"event": "train", "step": 6120, "loss": 3.6055747628211976, "learning_rate": 0.0002827637813026663, "elapsed_s": 158.10142922401428}
+{"event": "train", "step": 6140, "loss": 3.631784987449646, "learning_rate": 0.0002815188690990979, "elapsed_s": 158.55351638793945}
+{"event": "train", "step": 6160, "loss": 3.5908008813858032, "learning_rate": 0.00028027316218940193, "elapsed_s": 159.00856733322144}
+{"event": "train", "step": 6180, "loss": 3.6100397229194643, "learning_rate": 0.0002790266919824121, "elapsed_s": 159.4791603088379}
+{"event": "train", "step": 6200, "loss": 3.611032509803772, "learning_rate": 0.00027777948990620737, "elapsed_s": 159.9340124130249}
+{"event": "train", "step": 6220, "loss": 3.5865638971328737, "learning_rate": 0.00027653158740732005, "elapsed_s": 160.40818452835083}
+{"event": "train", "step": 6240, "loss": 3.6482472896575926, "learning_rate": 0.0002752830159499424, "elapsed_s": 160.89025902748108}
+{"event": "train", "step": 6260, "loss": 3.6201611042022703, "learning_rate": 0.00027403380701513405, "elapsed_s": 161.34688925743103}
+{"event": "train", "step": 6280, "loss": 3.6529981970787047, "learning_rate": 0.00027278399210002715, "elapsed_s": 161.81778573989868}
+{"event": "train", "step": 6300, "loss": 3.6409942865371705, "learning_rate": 0.00027153360271703326, "elapsed_s": 162.27757143974304}
+{"event": "train", "step": 6320, "loss": 3.599995720386505, "learning_rate": 0.0002702826703930483, "elapsed_s": 162.73640847206116}
+{"event": "train", "step": 6340, "loss": 3.5940600633621216, "learning_rate": 0.0002690312266686577, "elapsed_s": 163.19120693206787}
+{"event": "train", "step": 6360, "loss": 3.598197567462921, "learning_rate": 0.0002677793030973411, "elapsed_s": 163.6658034324646}
+{"event": "train", "step": 6380, "loss": 3.6393325686454774, "learning_rate": 0.0002665269312446768, "elapsed_s": 164.12490010261536}
+{"event": "train", "step": 6400, "loss": 3.6001453399658203, "learning_rate": 0.00026527414268754615, "elapsed_s": 164.59718441963196}
+{"event": "train", "step": 6420, "loss": 3.624724876880646, "learning_rate": 0.00026402096901333703, "elapsed_s": 165.09443950653076}
+{"event": "train", "step": 6440, "loss": 3.602348256111145, "learning_rate": 0.00026276744181914736, "elapsed_s": 165.5491669178009}
+{"event": "train", "step": 6460, "loss": 3.5895267724990845, "learning_rate": 0.00026151359271098885, "elapsed_s": 166.032653093338}
+{"event": "train", "step": 6480, "loss": 3.5456287622451783, "learning_rate": 0.00026025945330298954, "elapsed_s": 166.4867033958435}
+{"event": "train", "step": 6500, "loss": 3.525104296207428, "learning_rate": 0.0002590050552165975, "elapsed_s": 166.96087741851807}
+{"event": "train", "step": 6520, "loss": 3.51563755273819, "learning_rate": 0.00025775043007978245, "elapsed_s": 167.42357897758484}
+{"event": "train", "step": 6540, "loss": 3.5496943950653077, "learning_rate": 0.0002564956095262392, "elapsed_s": 167.89999437332153}
+{"event": "train", "step": 6560, "loss": 3.5653562426567076, "learning_rate": 0.0002552406251945897, "elapsed_s": 168.34778094291687}
+{"event": "train", "step": 6580, "loss": 3.57904714345932, "learning_rate": 0.0002539855087275853, "elapsed_s": 168.801207780838}
+{"event": "train", "step": 6600, "loss": 3.566093146800995, "learning_rate": 0.000252730291771309, "elapsed_s": 169.26766800880432}
+{"event": "train", "step": 6620, "loss": 3.53422189950943, "learning_rate": 0.0002514750059743775, "elapsed_s": 169.7401888370514}
+{"event": "train", "step": 6640, "loss": 3.5672189354896546, "learning_rate": 0.00025021968298714317, "elapsed_s": 170.21868586540222}
+{"event": "train", "step": 6660, "loss": 3.560614824295044, "learning_rate": 0.0002489643544608962, "elapsed_s": 170.6764898300171}
+{"event": "train", "step": 6680, "loss": 3.58953697681427, "learning_rate": 0.0002477090520470662, "elapsed_s": 171.13046264648438}
+{"event": "train", "step": 6700, "loss": 3.564437139034271, "learning_rate": 0.0002464538073964247, "elapsed_s": 171.59273982048035}
+{"event": "train", "step": 6720, "loss": 3.5662947058677674, "learning_rate": 0.0002451986521582867, "elapsed_s": 172.05807638168335}
+{"event": "train", "step": 6740, "loss": 3.577035164833069, "learning_rate": 0.00024394361797971261, "elapsed_s": 172.52541851997375}
+{"event": "train", "step": 6760, "loss": 3.52260183095932, "learning_rate": 0.00024268873650471073, "elapsed_s": 173.00542426109314}
+{"event": "train", "step": 6780, "loss": 3.609348249435425, "learning_rate": 0.000241434039373439, "elapsed_s": 173.465083360672}
+{"event": "train", "step": 6800, "loss": 3.558931303024292, "learning_rate": 0.00024017955822140742, "elapsed_s": 173.9248127937317}
+{"event": "train", "step": 6820, "loss": 3.5746831059455872, "learning_rate": 0.00023892532467868028, "elapsed_s": 174.39372181892395}
+{"event": "train", "step": 6840, "loss": 3.560846281051636, "learning_rate": 0.00023767137036907892, "elapsed_s": 174.85144996643066}
+{"event": "train", "step": 6860, "loss": 3.5782679677009583, "learning_rate": 0.00023641772690938405, "elapsed_s": 175.32108855247498}
+{"event": "train", "step": 6880, "loss": 3.519067883491516, "learning_rate": 0.0002351644259085387, "elapsed_s": 175.76965284347534}
+{"event": "train", "step": 6900, "loss": 3.473187470436096, "learning_rate": 0.00023391149896685135, "elapsed_s": 176.28448104858398}
+{"event": "train", "step": 6920, "loss": 3.494801199436188, "learning_rate": 0.00023265897767519903, "elapsed_s": 176.72877645492554}
+{"event": "train", "step": 6940, "loss": 3.4969073057174684, "learning_rate": 0.00023140689361423084, "elapsed_s": 177.1912178993225}
+{"event": "train", "step": 6960, "loss": 3.4960999131202697, "learning_rate": 0.00023015527835357167, "elapsed_s": 177.64276099205017}
+{"event": "train", "step": 6980, "loss": 3.480712127685547, "learning_rate": 0.00022890416345102633, "elapsed_s": 178.08909511566162}
+{"event": "train", "step": 7000, "loss": 3.5234844207763674, "learning_rate": 0.00022765358045178367, "elapsed_s": 178.53773021697998}
+{"event": "validation", "step": 7000, "loss": 4.014070451259613, "recall_at_1": 0.0095}
+{"event": "train", "step": 7020, "loss": 3.5113906860351562, "learning_rate": 0.0002264035608876213, "elapsed_s": 179.30949926376343}
+{"event": "train", "step": 7040, "loss": 3.501691794395447, "learning_rate": 0.00022515413627611076, "elapsed_s": 179.78376936912537}
+{"event": "train", "step": 7060, "loss": 3.4988748669624328, "learning_rate": 0.0002239053381198224, "elapsed_s": 180.244934797287}
+{"event": "train", "step": 7080, "loss": 3.5081374764442446, "learning_rate": 0.00022265719790553147, "elapsed_s": 180.69414615631104}
+{"event": "train", "step": 7100, "loss": 3.5014626026153564, "learning_rate": 0.00022140974710342401, "elapsed_s": 181.15194749832153}
+{"event": "train", "step": 7120, "loss": 3.5529001355171204, "learning_rate": 0.00022016301716630367, "elapsed_s": 181.61757922172546}
+{"event": "train", "step": 7140, "loss": 3.5365555763244627, "learning_rate": 0.0002189170395287982, "elapsed_s": 182.0828993320465}
+{"event": "train", "step": 7160, "loss": 3.535981774330139, "learning_rate": 0.00021767184560656715, "elapsed_s": 182.53299498558044}
+{"event": "train", "step": 7180, "loss": 3.4913656949996947, "learning_rate": 0.00021642746679551, "elapsed_s": 182.98670434951782}
+{"event": "train", "step": 7200, "loss": 3.5157767057418825, "learning_rate": 0.00021518393447097402, "elapsed_s": 183.4324722290039}
+{"event": "train", "step": 7220, "loss": 3.478971564769745, "learning_rate": 0.00021394127998696348, "elapsed_s": 183.8839569091797}
+{"event": "train", "step": 7240, "loss": 3.4989853501319885, "learning_rate": 0.00021269953467534906, "elapsed_s": 184.33111095428467}
+{"event": "train", "step": 7260, "loss": 3.5408986806869507, "learning_rate": 0.00021145872984507818, "elapsed_s": 184.78522992134094}
+{"event": "train", "step": 7280, "loss": 3.522625720500946, "learning_rate": 0.0002102188967813849, "elapsed_s": 185.37367725372314}
+{"event": "train", "step": 7300, "loss": 3.5111509203910827, "learning_rate": 0.00020898006674500156, "elapsed_s": 185.82871675491333}
+{"event": "train", "step": 7320, "loss": 3.4920351266860963, "learning_rate": 0.00020774227097137074, "elapsed_s": 186.30879259109497}
+{"event": "train", "step": 7340, "loss": 3.414402890205383, "learning_rate": 0.0002065055406698573, "elapsed_s": 186.75121188163757}
+{"event": "train", "step": 7360, "loss": 3.420544457435608, "learning_rate": 0.00020526990702296154, "elapsed_s": 187.20256638526917}
+{"event": "train", "step": 7380, "loss": 3.4592064023017883, "learning_rate": 0.00020403540118553315, "elapsed_s": 187.67271375656128}
+{"event": "train", "step": 7400, "loss": 3.457471692562103, "learning_rate": 0.00020280205428398584, "elapsed_s": 188.15197277069092}
+{"event": "train", "step": 7420, "loss": 3.4199041724205017, "learning_rate": 0.0002015698974155121, "elapsed_s": 188.6072542667389}
+{"event": "train", "step": 7440, "loss": 3.433551073074341, "learning_rate": 0.00020033896164729908, "elapsed_s": 189.07893252372742}
+{"event": "train", "step": 7460, "loss": 3.450441634654999, "learning_rate": 0.00019910927801574613, "elapsed_s": 189.589173078537}
+{"event": "train", "step": 7480, "loss": 3.446766746044159, "learning_rate": 0.00019788087752568116, "elapsed_s": 190.07831358909607}
+{"event": "train", "step": 7500, "loss": 3.4861266136169435, "learning_rate": 0.00019665379114957955, "elapsed_s": 190.53027987480164}
+{"event": "train", "step": 7520, "loss": 3.4873907446861265, "learning_rate": 0.00019542804982678285, "elapsed_s": 190.97902083396912}
+{"event": "train", "step": 7540, "loss": 3.461447906494141, "learning_rate": 0.00019420368446271936, "elapsed_s": 191.43285393714905}
+{"event": "train", "step": 7560, "loss": 3.4535791993141176, "learning_rate": 0.000192980725928124, "elapsed_s": 191.90197205543518}
+{"event": "train", "step": 7580, "loss": 3.443385970592499, "learning_rate": 0.00019175920505826044, "elapsed_s": 192.3638973236084}
+{"event": "train", "step": 7600, "loss": 3.4891810774803163, "learning_rate": 0.00019053915265214376, "elapsed_s": 192.8144335746765}
+{"event": "train", "step": 7620, "loss": 3.4586902618408204, "learning_rate": 0.00018932059947176345, "elapsed_s": 193.26861429214478}
+{"event": "train", "step": 7640, "loss": 3.4931727766990663, "learning_rate": 0.00018810357624130813, "elapsed_s": 193.72420167922974}
+{"event": "train", "step": 7660, "loss": 3.4230407953262327, "learning_rate": 0.0001868881136463908, "elapsed_s": 194.1904571056366}
+{"event": "train", "step": 7680, "loss": 3.449254775047302, "learning_rate": 0.0001856742423332751, "elapsed_s": 194.64990997314453}
+{"event": "train", "step": 7700, "loss": 3.460995650291443, "learning_rate": 0.00018446199290810263, "elapsed_s": 195.10585403442383}
+{"event": "train", "step": 7720, "loss": 3.4485567450523376, "learning_rate": 0.00018325139593612118, "elapsed_s": 195.56692838668823}
+{"event": "train", "step": 7740, "loss": 3.4507665634155273, "learning_rate": 0.00018204248194091428, "elapsed_s": 196.01351690292358}
+{"event": "train", "step": 7760, "loss": 3.3553954124450684, "learning_rate": 0.00018083528140363142, "elapsed_s": 196.4935052394867}
+{"event": "train", "step": 7780, "loss": 3.3693652510643006, "learning_rate": 0.00017962982476221946, "elapsed_s": 196.93833255767822}
+{"event": "train", "step": 7800, "loss": 3.3778060913085937, "learning_rate": 0.0001784261424106552, "elapsed_s": 197.3886318206787}
+{"event": "train", "step": 7820, "loss": 3.3765132069587707, "learning_rate": 0.0001772242646981794, "elapsed_s": 197.86250591278076}
+{"event": "train", "step": 7840, "loss": 3.427934718132019, "learning_rate": 0.00017602422192853094, "elapsed_s": 198.3071310520172}
+{"event": "train", "step": 7860, "loss": 3.3832286596298218, "learning_rate": 0.00017482604435918315, "elapsed_s": 198.7598385810852}
+{"event": "train", "step": 7880, "loss": 3.4097873091697695, "learning_rate": 0.00017362976220058093, "elapsed_s": 199.20663285255432}
+{"event": "train", "step": 7900, "loss": 3.3625258445739745, "learning_rate": 0.0001724354056153789, "elapsed_s": 199.68357038497925}
+{"event": "train", "step": 7920, "loss": 3.397590160369873, "learning_rate": 0.00017124300471768085, "elapsed_s": 200.14575791358948}
+{"event": "train", "step": 7940, "loss": 3.37775536775589, "learning_rate": 0.00017005258957228047, "elapsed_s": 200.59791564941406}
+{"event": "train", "step": 7960, "loss": 3.3868757367134092, "learning_rate": 0.00016886419019390364, "elapsed_s": 201.0479371547699}
+{"event": "train", "step": 7980, "loss": 3.3824564933776857, "learning_rate": 0.00016767783654645114, "elapsed_s": 201.5015082359314}
+{"event": "train", "step": 8000, "loss": 3.405401384830475, "learning_rate": 0.00016649355854224336, "elapsed_s": 201.97464227676392}
+{"event": "validation", "step": 8000, "loss": 4.028741419315338, "recall_at_1": 0.013}
+{"event": "train", "step": 8020, "loss": 3.4034392833709717, "learning_rate": 0.00016531138604126616, "elapsed_s": 202.8186318874359}
+{"event": "train", "step": 8040, "loss": 3.414106845855713, "learning_rate": 0.0001641313488504181, "elapsed_s": 203.29016041755676}
+{"event": "train", "step": 8060, "loss": 3.420532500743866, "learning_rate": 0.00016295347672275862, "elapsed_s": 203.75265622138977}
+{"event": "train", "step": 8080, "loss": 3.4187070727348328, "learning_rate": 0.00016177779935675792, "elapsed_s": 204.20082139968872}
+{"event": "train", "step": 8100, "loss": 3.443345141410828, "learning_rate": 0.00016060434639554835, "elapsed_s": 204.64709043502808}
+{"event": "train", "step": 8120, "loss": 3.416650629043579, "learning_rate": 0.00015943314742617687, "elapsed_s": 205.11098074913025}
+{"event": "train", "step": 8140, "loss": 3.407922625541687, "learning_rate": 0.00015826423197885895, "elapsed_s": 205.56686329841614}
+{"event": "train", "step": 8160, "loss": 3.4033721446990968, "learning_rate": 0.0001570976295262341, "elapsed_s": 206.02381038665771}
+{"event": "train", "step": 8180, "loss": 3.3508713960647585, "learning_rate": 0.00015593336948262296, "elapsed_s": 206.4827859401703}
+{"event": "train", "step": 8200, "loss": 3.3264158606529235, "learning_rate": 0.00015477148120328525, "elapsed_s": 206.94370126724243}
+{"event": "train", "step": 8220, "loss": 3.357633078098297, "learning_rate": 0.00015361199398367976, "elapsed_s": 207.40177273750305}
+{"event": "train", "step": 8240, "loss": 3.334688150882721, "learning_rate": 0.00015245493705872621, "elapsed_s": 207.85056495666504}
+{"event": "train", "step": 8260, "loss": 3.3416346311569214, "learning_rate": 0.0001513003396020672, "elapsed_s": 208.3019335269928}
+{"event": "train", "step": 8280, "loss": 3.359859812259674, "learning_rate": 0.00015014823072533327, "elapsed_s": 208.74837803840637}
+{"event": "train", "step": 8300, "loss": 3.328729438781738, "learning_rate": 0.00014899863947740878, "elapsed_s": 209.19875741004944}
+{"event": "train", "step": 8320, "loss": 3.2802744030952455, "learning_rate": 0.00014785159484369964, "elapsed_s": 209.66893219947815}
+{"event": "train", "step": 8340, "loss": 3.325623834133148, "learning_rate": 0.00014670712574540198, "elapsed_s": 210.13180541992188}
+{"event": "train", "step": 8360, "loss": 3.380221259593964, "learning_rate": 0.00014556526103877326, "elapsed_s": 210.59650993347168}
+{"event": "train", "step": 8380, "loss": 3.310933494567871, "learning_rate": 0.00014442602951440507, "elapsed_s": 211.0432093143463}
+{"event": "train", "step": 8400, "loss": 3.3468007802963258, "learning_rate": 0.00014328945989649638, "elapsed_s": 211.49689722061157}
+{"event": "train", "step": 8420, "loss": 3.3335564970970153, "learning_rate": 0.00014215558084213, "elapsed_s": 211.9429337978363}
+{"event": "train", "step": 8440, "loss": 3.3511354327201843, "learning_rate": 0.00014102442094054968, "elapsed_s": 212.40075969696045}
+{"event": "train", "step": 8460, "loss": 3.350973975658417, "learning_rate": 0.00013989600871243938, "elapsed_s": 212.84903740882874}
+{"event": "train", "step": 8480, "loss": 3.342862379550934, "learning_rate": 0.0001387703726092042, "elapsed_s": 213.31339764595032}
+{"event": "train", "step": 8500, "loss": 3.3313992261886596, "learning_rate": 0.00013764754101225287, "elapsed_s": 213.76490664482117}
+{"event": "train", "step": 8520, "loss": 3.3642133235931397, "learning_rate": 0.00013652754223228243, "elapsed_s": 214.21392369270325}
+{"event": "train", "step": 8540, "loss": 3.3833843946456907, "learning_rate": 0.00013541040450856402, "elapsed_s": 214.66894960403442}
+{"event": "train", "step": 8560, "loss": 3.3519205331802366, "learning_rate": 0.0001342961560082312, "elapsed_s": 215.12941527366638}
+{"event": "train", "step": 8580, "loss": 3.3570930480957033, "learning_rate": 0.00013318482482556958, "elapsed_s": 215.59551811218262}
+{"event": "train", "step": 8600, "loss": 3.358628273010254, "learning_rate": 0.00013207643898130854, "elapsed_s": 216.0326418876648}
+{"event": "train", "step": 8620, "loss": 3.2622212409973144, "learning_rate": 0.0001309710264219146, "elapsed_s": 216.4982147216797}
+{"event": "train", "step": 8640, "loss": 3.24926997423172, "learning_rate": 0.00012986861501888686, "elapsed_s": 216.94762325286865}
+{"event": "train", "step": 8660, "loss": 3.2485180735588073, "learning_rate": 0.00012876923256805456, "elapsed_s": 217.4028832912445}
+{"event": "train", "step": 8680, "loss": 3.2541823625564574, "learning_rate": 0.0001276729067888757, "elapsed_s": 217.84329962730408}
+{"event": "train", "step": 8700, "loss": 3.294660413265228, "learning_rate": 0.00012657966532373847, "elapsed_s": 218.28474354743958}
+{"event": "train", "step": 8720, "loss": 3.2748775839805604, "learning_rate": 0.00012548953573726413, "elapsed_s": 218.73736572265625}
+{"event": "train", "step": 8740, "loss": 3.2837973952293398, "learning_rate": 0.00012440254551561217, "elapsed_s": 219.22505140304565}
+{"event": "train", "step": 8760, "loss": 3.2854265928268434, "learning_rate": 0.00012331872206578715, "elapsed_s": 219.6735384464264}
+{"event": "train", "step": 8780, "loss": 3.28590270280838, "learning_rate": 0.0001222380927149477, "elapsed_s": 220.1216242313385}
+{"event": "train", "step": 8800, "loss": 3.2714962005615233, "learning_rate": 0.00012116068470971763, "elapsed_s": 220.56429290771484}
+{"event": "train", "step": 8820, "loss": 3.2879531621932983, "learning_rate": 0.00012008652521549877, "elapsed_s": 221.0680797100067}
+{"event": "train", "step": 8840, "loss": 3.313975524902344, "learning_rate": 0.00011901564131578598, "elapsed_s": 221.51140451431274}
+{"event": "train", "step": 8860, "loss": 3.287469041347504, "learning_rate": 0.00011794806001148442, "elapsed_s": 221.9614691734314}
+{"event": "train", "step": 8880, "loss": 3.3123512744903563, "learning_rate": 0.00011688380822022906, "elapsed_s": 222.40334343910217}
+{"event": "train", "step": 8900, "loss": 3.2921712398529053, "learning_rate": 0.00011582291277570508, "elapsed_s": 222.84673738479614}
+{"event": "train", "step": 8920, "loss": 3.2859559178352358, "learning_rate": 0.00011476540042697217, "elapsed_s": 223.28989934921265}
+{"event": "train", "step": 8940, "loss": 3.3041190981864927, "learning_rate": 0.00011371129783778994, "elapsed_s": 223.74363160133362}
+{"event": "train", "step": 8960, "loss": 3.282928478717804, "learning_rate": 0.0001126606315859452, "elapsed_s": 224.1956913471222}
+{"event": "train", "step": 8980, "loss": 3.2851489067077635, "learning_rate": 0.00011161342816258224, "elapsed_s": 224.7834403514862}
+{"event": "train", "step": 9000, "loss": 3.3162544846534727, "learning_rate": 0.00011056971397153462, "elapsed_s": 225.22769594192505}
+{"event": "validation", "step": 9000, "loss": 4.0315704345703125, "recall_at_1": 0.014}
+{"event": "train", "step": 9020, "loss": 3.315142738819122, "learning_rate": 0.00010952951532865987, "elapsed_s": 225.91105103492737}
+{"event": "train", "step": 9040, "loss": 3.2729999303817747, "learning_rate": 0.00010849285846117543, "elapsed_s": 226.37458992004395}
+{"event": "train", "step": 9060, "loss": 3.2146494507789614, "learning_rate": 0.00010745976950699765, "elapsed_s": 226.83263659477234}
+{"event": "train", "step": 9080, "loss": 3.236351466178894, "learning_rate": 0.00010643027451408274, "elapsed_s": 227.28238701820374}
+{"event": "train", "step": 9100, "loss": 3.218580424785614, "learning_rate": 0.00010540439943977, "elapsed_s": 227.7251856327057}
+{"event": "train", "step": 9120, "loss": 3.226340413093567, "learning_rate": 0.00010438217015012735, "elapsed_s": 228.1758484840393}
+{"event": "train", "step": 9140, "loss": 3.2384204626083375, "learning_rate": 0.00010336361241929893, "elapsed_s": 228.63273549079895}
+{"event": "train", "step": 9160, "loss": 3.209488832950592, "learning_rate": 0.00010234875192885576, "elapsed_s": 229.07627201080322}
+{"event": "train", "step": 9180, "loss": 3.2223552107810973, "learning_rate": 0.00010133761426714771, "elapsed_s": 229.51678705215454}
+{"event": "train", "step": 9200, "loss": 3.2277907371520995, "learning_rate": 0.00010033022492865846, "elapsed_s": 229.96678066253662}
+{"event": "train", "step": 9220, "loss": 3.253125751018524, "learning_rate": 9.932660931336279e-05, "elapsed_s": 230.42082142829895}
+{"event": "train", "step": 9240, "loss": 3.2035163283348083, "learning_rate": 9.832679272608607e-05, "elapsed_s": 230.86504483222961}
+{"event": "train", "step": 9260, "loss": 3.2087109088897705, "learning_rate": 9.733080037586636e-05, "elapsed_s": 231.30664443969727}
+{"event": "train", "step": 9280, "loss": 3.2302934169769286, "learning_rate": 9.633865737531847e-05, "elapsed_s": 231.75756120681763}
+{"event": "train", "step": 9300, "loss": 3.259160947799683, "learning_rate": 9.535038874000134e-05, "elapsed_s": 232.19866609573364}
+{"event": "train", "step": 9320, "loss": 3.20722553730011, "learning_rate": 9.436601938778672e-05, "elapsed_s": 232.65176677703857}
+{"event": "train", "step": 9340, "loss": 3.2426738381385802, "learning_rate": 9.338557413823123e-05, "elapsed_s": 233.09904885292053}
+{"event": "train", "step": 9360, "loss": 3.243328070640564, "learning_rate": 9.240907771195046e-05, "elapsed_s": 233.54139018058777}
+{"event": "train", "step": 9380, "loss": 3.2323155879974363, "learning_rate": 9.143655472999571e-05, "elapsed_s": 233.99250149726868}
+{"event": "train", "step": 9400, "loss": 3.2595462560653687, "learning_rate": 9.046802971323323e-05, "elapsed_s": 234.43108296394348}
+{"event": "train", "step": 9420, "loss": 3.241642379760742, "learning_rate": 8.950352708172583e-05, "elapsed_s": 234.88102293014526}
+{"event": "train", "step": 9440, "loss": 3.2118648171424864, "learning_rate": 8.85430711541175e-05, "elapsed_s": 235.32006168365479}
+{"event": "train", "step": 9460, "loss": 3.2656757116317747, "learning_rate": 8.758668614701973e-05, "elapsed_s": 235.74381566047668}
+{"event": "train", "step": 9480, "loss": 3.1507972717285155, "learning_rate": 8.663439617440133e-05, "elapsed_s": 236.21121859550476}
+{"event": "train", "step": 9500, "loss": 3.1775978684425352, "learning_rate": 8.568622524698025e-05, "elapsed_s": 236.68364238739014}
+{"event": "train", "step": 9520, "loss": 3.151635706424713, "learning_rate": 8.474219727161823e-05, "elapsed_s": 237.12621688842773}
+{"event": "train", "step": 9540, "loss": 3.207791471481323, "learning_rate": 8.3802336050718e-05, "elapsed_s": 237.56465792655945}
+{"event": "train", "step": 9560, "loss": 3.1998273372650146, "learning_rate": 8.286666528162312e-05, "elapsed_s": 238.0118863582611}
+{"event": "train", "step": 9580, "loss": 3.2005961775779723, "learning_rate": 8.193520855602071e-05, "elapsed_s": 238.4600419998169}
+{"event": "train", "step": 9600, "loss": 3.1647737503051756, "learning_rate": 8.100798935934622e-05, "elapsed_s": 238.89878010749817}
+{"event": "train", "step": 9620, "loss": 3.1894432663917542, "learning_rate": 8.008503107019152e-05, "elapsed_s": 239.33746004104614}
+{"event": "train", "step": 9640, "loss": 3.1581084609031675, "learning_rate": 7.916635695971539e-05, "elapsed_s": 239.78419852256775}
+{"event": "train", "step": 9660, "loss": 3.1616514801979063, "learning_rate": 7.825199019105677e-05, "elapsed_s": 240.22793674468994}
+{"event": "train", "step": 9680, "loss": 3.1746445298194885, "learning_rate": 7.73419538187508e-05, "elapsed_s": 240.67883801460266}
+{"event": "train", "step": 9700, "loss": 3.2254854440689087, "learning_rate": 7.643627078814733e-05, "elapsed_s": 241.1315155029297}
+{"event": "train", "step": 9720, "loss": 3.1785324215888977, "learning_rate": 7.553496393483278e-05, "elapsed_s": 241.57239866256714}
+{"event": "train", "step": 9740, "loss": 3.204308795928955, "learning_rate": 7.463805598405384e-05, "elapsed_s": 242.00976610183716}
+{"event": "train", "step": 9760, "loss": 3.2123955845832826, "learning_rate": 7.374556955014491e-05, "elapsed_s": 242.44705319404602}
+{"event": "train", "step": 9780, "loss": 3.1901230573654176, "learning_rate": 7.285752713595762e-05, "elapsed_s": 242.8946976661682}
+{"event": "train", "step": 9800, "loss": 3.2202120304107664, "learning_rate": 7.197395113229388e-05, "elapsed_s": 243.34378576278687}
+{"event": "train", "step": 9820, "loss": 3.141201674938202, "learning_rate": 7.109486381734062e-05, "elapsed_s": 243.78167724609375}
+{"event": "train", "step": 9840, "loss": 3.1738998889923096, "learning_rate": 7.022028735610869e-05, "elapsed_s": 244.2193615436554}
+{"event": "train", "step": 9860, "loss": 3.1878626942634583, "learning_rate": 6.935024379987401e-05, "elapsed_s": 244.65535640716553}
+{"event": "train", "step": 9880, "loss": 3.2226706027984617, "learning_rate": 6.848475508562103e-05, "elapsed_s": 245.1013069152832}
+{"event": "train", "step": 9900, "loss": 3.150554049015045, "learning_rate": 6.762384303549005e-05, "elapsed_s": 245.55476832389832}
+{"event": "train", "step": 9920, "loss": 3.1035247087478637, "learning_rate": 6.676752935622694e-05, "elapsed_s": 245.99321103096008}
+{"event": "train", "step": 9940, "loss": 3.1464468955993654, "learning_rate": 6.591583563863588e-05, "elapsed_s": 246.4295973777771}
+{"event": "train", "step": 9960, "loss": 3.109871971607208, "learning_rate": 6.506878335703483e-05, "elapsed_s": 246.868803024292}
+{"event": "train", "step": 9980, "loss": 3.1072550058364867, "learning_rate": 6.422639386871393e-05, "elapsed_s": 247.3212013244629}
+{"event": "train", "step": 10000, "loss": 3.1230887532234193, "learning_rate": 6.338868841339763e-05, "elapsed_s": 247.7740375995636}
+{"event": "validation", "step": 10000, "loss": 4.059190899133682, "recall_at_1": 0.012}
+{"event": "train", "step": 10020, "loss": 3.1350090146064757, "learning_rate": 6.255568811270848e-05, "elapsed_s": 248.49190998077393}
+{"event": "train", "step": 10040, "loss": 3.133158195018768, "learning_rate": 6.172741396963488e-05, "elapsed_s": 248.95368456840515}
+{"event": "train", "step": 10060, "loss": 3.149695360660553, "learning_rate": 6.090388686800147e-05, "elapsed_s": 249.40679240226746}
+{"event": "train", "step": 10080, "loss": 3.1139381647109987, "learning_rate": 6.0085127571942815e-05, "elapsed_s": 249.84389519691467}
+{"event": "train", "step": 10100, "loss": 3.167892634868622, "learning_rate": 5.927115672537936e-05, "elapsed_s": 250.28086471557617}
+{"event": "train", "step": 10120, "loss": 3.163481831550598, "learning_rate": 5.8461994851497306e-05, "elapsed_s": 250.7307152748108}
+{"event": "train", "step": 10140, "loss": 3.1451456427574156, "learning_rate": 5.765766235223108e-05, "elapsed_s": 251.1844916343689}
+{"event": "train", "step": 10160, "loss": 3.1567931890487673, "learning_rate": 5.68581795077488e-05, "elapsed_s": 251.6237597465515}
+{"event": "train", "step": 10180, "loss": 3.1214336156845093, "learning_rate": 5.60635664759411e-05, "elapsed_s": 252.05971503257751}
+{"event": "train", "step": 10200, "loss": 3.152518653869629, "learning_rate": 5.5273843291912704e-05, "elapsed_s": 252.50494718551636}
+{"event": "train", "step": 10220, "loss": 3.1787641763687136, "learning_rate": 5.448902986747756e-05, "elapsed_s": 252.95064973831177}
+{"event": "train", "step": 10240, "loss": 3.179734635353088, "learning_rate": 5.37091459906564e-05, "elapsed_s": 253.4127540588379}
+{"event": "train", "step": 10260, "loss": 3.1562291026115417, "learning_rate": 5.293421132517809e-05, "elapsed_s": 253.87357234954834}
+{"event": "train", "step": 10280, "loss": 3.1514843583106993, "learning_rate": 5.216424540998374e-05, "elapsed_s": 254.33766627311707}
+{"event": "train", "step": 10300, "loss": 3.158307433128357, "learning_rate": 5.139926765873404e-05, "elapsed_s": 254.8212697505951}
+{"event": "train", "step": 10320, "loss": 3.142444670200348, "learning_rate": 5.063929735931985e-05, "elapsed_s": 255.26701498031616}
+{"event": "train", "step": 10340, "loss": 3.1055340647697447, "learning_rate": 4.988435367337571e-05, "elapsed_s": 255.72527027130127}
+{"event": "train", "step": 10360, "loss": 3.053577220439911, "learning_rate": 4.913445563579705e-05, "elapsed_s": 256.16317892074585}
+{"event": "train", "step": 10380, "loss": 3.0902682185173034, "learning_rate": 4.8389622154259874e-05, "elapsed_s": 256.60102462768555}
+{"event": "train", "step": 10400, "loss": 3.099741744995117, "learning_rate": 4.764987200874413e-05, "elapsed_s": 257.0561625957489}
+{"event": "train", "step": 10420, "loss": 3.0978921175003054, "learning_rate": 4.691522385106034e-05, "elapsed_s": 257.4925172328949}
+{"event": "train", "step": 10440, "loss": 3.0883382320404054, "learning_rate": 4.618569620437921e-05, "elapsed_s": 257.931049823761}
+{"event": "train", "step": 10460, "loss": 3.107429313659668, "learning_rate": 4.546130746276461e-05, "elapsed_s": 258.4021770954132}
+{"event": "train", "step": 10480, "loss": 3.0427208065986635, "learning_rate": 4.474207589070969e-05, "elapsed_s": 258.85897517204285}
+{"event": "train", "step": 10500, "loss": 3.084516775608063, "learning_rate": 4.402801962267669e-05, "elapsed_s": 259.3011951446533}
+{"event": "train", "step": 10520, "loss": 3.0976978182792663, "learning_rate": 4.3319156662639306e-05, "elapsed_s": 259.74048137664795}
+{"event": "train", "step": 10540, "loss": 3.1088712573051454, "learning_rate": 4.261550488362895e-05, "elapsed_s": 260.18141412734985}
+{"event": "train", "step": 10560, "loss": 3.1005751729011535, "learning_rate": 4.191708202728411e-05, "elapsed_s": 260.6310076713562}
+{"event": "train", "step": 10580, "loss": 3.101475739479065, "learning_rate": 4.122390570340292e-05, "elapsed_s": 261.08052015304565}
+{"event": "train", "step": 10600, "loss": 3.068959641456604, "learning_rate": 4.0535993389499285e-05, "elapsed_s": 261.54053807258606}
+{"event": "train", "step": 10620, "loss": 3.1538814187049864, "learning_rate": 3.9853362430362055e-05, "elapsed_s": 261.98338317871094}
+{"event": "train", "step": 10640, "loss": 3.1186363816261293, "learning_rate": 3.917603003761797e-05, "elapsed_s": 262.42552733421326}
+{"event": "train", "step": 10660, "loss": 3.0991116881370546, "learning_rate": 3.850401328929726e-05, "elapsed_s": 262.86816024780273}
+{"event": "train", "step": 10680, "loss": 3.12226482629776, "learning_rate": 3.783732912940341e-05, "elapsed_s": 263.424161195755}
+{"event": "train", "step": 10700, "loss": 3.1158600091934203, "learning_rate": 3.717599436748595e-05, "elapsed_s": 263.8786907196045}
+{"event": "train", "step": 10720, "loss": 3.117265856266022, "learning_rate": 3.652002567821619e-05, "elapsed_s": 264.32139015197754}
+{"event": "train", "step": 10740, "loss": 3.120661723613739, "learning_rate": 3.586943960096731e-05, "elapsed_s": 264.7825720310211}
+{"event": "train", "step": 10760, "loss": 3.0678571105003356, "learning_rate": 3.522425253939707e-05, "elapsed_s": 265.2350196838379}
+{"event": "train", "step": 10780, "loss": 3.052863264083862, "learning_rate": 3.4584480761034376e-05, "elapsed_s": 265.67457938194275}
+{"event": "train", "step": 10800, "loss": 3.0339601397514344, "learning_rate": 3.395014039686883e-05, "elapsed_s": 266.12393403053284}
+{"event": "train", "step": 10820, "loss": 3.0815529346466066, "learning_rate": 3.3321247440944336e-05, "elapsed_s": 266.56317257881165}
+{"event": "train", "step": 10840, "loss": 3.0721697330474855, "learning_rate": 3.269781774995551e-05, "elapsed_s": 267.0167965888977}
+{"event": "train", "step": 10860, "loss": 3.0721855044364927, "learning_rate": 3.2079867042848375e-05, "elapsed_s": 267.460590839386}
+{"event": "train", "step": 10880, "loss": 3.066568922996521, "learning_rate": 3.146741090042332e-05, "elapsed_s": 267.90470600128174}
+{"event": "train", "step": 10900, "loss": 3.0655528903007507, "learning_rate": 3.0860464764942805e-05, "elapsed_s": 268.36712312698364}
+{"event": "train", "step": 10920, "loss": 3.0688084959983826, "learning_rate": 3.0259043939741964e-05, "elapsed_s": 268.81864619255066}
+{"event": "train", "step": 10940, "loss": 3.076883053779602, "learning_rate": 2.9663163588842414e-05, "elapsed_s": 269.27081823349}
+{"event": "train", "step": 10960, "loss": 3.071745145320892, "learning_rate": 2.907283873657024e-05, "elapsed_s": 269.71284079551697}
+{"event": "train", "step": 10980, "loss": 3.0473270773887635, "learning_rate": 2.8488084267176945e-05, "elapsed_s": 270.15330934524536}
+{"event": "train", "step": 11000, "loss": 3.0632264256477355, "learning_rate": 2.7908914924464486e-05, "elapsed_s": 270.59177470207214}
+{"event": "validation", "step": 11000, "loss": 4.100206524133682, "recall_at_1": 0.0125}
+{"event": "train", "step": 11020, "loss": 3.055523455142975, "learning_rate": 2.7335345311413183e-05, "elapsed_s": 271.33377599716187}
+{"event": "train", "step": 11040, "loss": 3.063327395915985, "learning_rate": 2.6767389889813502e-05, "elapsed_s": 271.7792775630951}
+{"event": "train", "step": 11060, "loss": 3.073573887348175, "learning_rate": 2.6205062979901968e-05, "elapsed_s": 272.2214095592499}
+{"event": "train", "step": 11080, "loss": 3.062827777862549, "learning_rate": 2.564837875999937e-05, "elapsed_s": 272.672963142395}
+{"event": "train", "step": 11100, "loss": 3.053048312664032, "learning_rate": 2.5097351266153785e-05, "elapsed_s": 273.12441968917847}
+{"event": "train", "step": 11120, "loss": 3.0643291592597963, "learning_rate": 2.4551994391786404e-05, "elapsed_s": 273.56644463539124}
+{"event": "train", "step": 11140, "loss": 3.087888014316559, "learning_rate": 2.401232188734151e-05, "elapsed_s": 274.0098285675049}
+{"event": "train", "step": 11160, "loss": 3.082514536380768, "learning_rate": 2.347834735993942e-05, "elapsed_s": 274.4572048187256}
+{"event": "train", "step": 11180, "loss": 3.103651964664459, "learning_rate": 2.2950084273033632e-05, "elapsed_s": 274.8897211551666}
+{"event": "train", "step": 11200, "loss": 3.042043697834015, "learning_rate": 2.2427545946071354e-05, "elapsed_s": 275.3593327999115}
+{"event": "train", "step": 11220, "loss": 3.034139370918274, "learning_rate": 2.1910745554157563e-05, "elapsed_s": 275.800208568573}
+{"event": "train", "step": 11240, "loss": 3.0457674860954285, "learning_rate": 2.1399696127722967e-05, "elapsed_s": 276.2392621040344}
+{"event": "train", "step": 11260, "loss": 3.017963409423828, "learning_rate": 2.0894410552195237e-05, "elapsed_s": 276.6792905330658}
+{"event": "train", "step": 11280, "loss": 3.036688816547394, "learning_rate": 2.039490156767443e-05, "elapsed_s": 277.12868428230286}
+{"event": "train", "step": 11300, "loss": 3.0644516944885254, "learning_rate": 1.990118176861147e-05, "elapsed_s": 277.58616852760315}
+{"event": "train", "step": 11320, "loss": 3.024069273471832, "learning_rate": 1.9413263603490688e-05, "elapsed_s": 278.0390646457672}
+{"event": "train", "step": 11340, "loss": 3.0538419604301454, "learning_rate": 1.893115937451609e-05, "elapsed_s": 278.4787492752075}
+{"event": "train", "step": 11360, "loss": 3.0692208409309387, "learning_rate": 1.8454881237300935e-05, "elapsed_s": 278.9313199520111}
+{"event": "train", "step": 11380, "loss": 3.043477237224579, "learning_rate": 1.798444120056142e-05, "elapsed_s": 279.3802180290222}
+{"event": "train", "step": 11400, "loss": 3.0229502439498903, "learning_rate": 1.7519851125813863e-05, "elapsed_s": 279.82037830352783}
+{"event": "train", "step": 11420, "loss": 3.0357467889785767, "learning_rate": 1.7061122727075662e-05, "elapsed_s": 280.26121711730957}
+{"event": "train", "step": 11440, "loss": 3.0675564050674438, "learning_rate": 1.660826757056977e-05, "elapsed_s": 280.7016849517822}
+{"event": "train", "step": 11460, "loss": 3.0249022245407104, "learning_rate": 1.6161297074433284e-05, "elapsed_s": 281.1628451347351}
+{"event": "train", "step": 11480, "loss": 3.0109732151031494, "learning_rate": 1.5720222508429442e-05, "elapsed_s": 281.6128840446472}
+{"event": "train", "step": 11500, "loss": 3.0231672644615175, "learning_rate": 1.5285054993663506e-05, "elapsed_s": 282.05439853668213}
+{"event": "train", "step": 11520, "loss": 3.0296661376953127, "learning_rate": 1.4855805502302255e-05, "elapsed_s": 282.49338126182556}
+{"event": "train", "step": 11540, "loss": 3.053460478782654, "learning_rate": 1.4432484857297507e-05, "elapsed_s": 282.9329240322113}
+{"event": "train", "step": 11560, "loss": 3.0547017574310305, "learning_rate": 1.4015103732113177e-05, "elapsed_s": 283.39268684387207}
+{"event": "train", "step": 11580, "loss": 3.0302084922790526, "learning_rate": 1.3603672650456062e-05, "elapsed_s": 283.8337211608887}
+{"event": "train", "step": 11600, "loss": 3.013373684883118, "learning_rate": 1.3198201986010572e-05, "elapsed_s": 284.2757830619812}
+{"event": "train", "step": 11620, "loss": 3.044739067554474, "learning_rate": 1.2798701962177223e-05, "elapsed_s": 284.721271276474}
+{"event": "train", "step": 11640, "loss": 3.0162797212600707, "learning_rate": 1.2405182651814755e-05, "elapsed_s": 285.17738032341003}
+{"event": "train", "step": 11660, "loss": 3.0055045962333677, "learning_rate": 1.201765397698626e-05, "elapsed_s": 285.6132996082306}
+{"event": "train", "step": 11680, "loss": 3.029694414138794, "learning_rate": 1.1636125708708895e-05, "elapsed_s": 286.0571508407593}
+{"event": "train", "step": 11700, "loss": 3.031453776359558, "learning_rate": 1.1260607466707733e-05, "elapsed_s": 286.4968707561493}
+{"event": "train", "step": 11720, "loss": 3.0268878698349, "learning_rate": 1.0891108719172926e-05, "elapsed_s": 286.95616602897644}
+{"event": "train", "step": 11740, "loss": 3.0082726001739504, "learning_rate": 1.0527638782521193e-05, "elapsed_s": 287.4127531051636}
+{"event": "train", "step": 11760, "loss": 2.9949403405189514, "learning_rate": 1.0170206821160905e-05, "elapsed_s": 287.8803381919861}
+{"event": "train", "step": 11780, "loss": 3.0258309483528136, "learning_rate": 9.818821847260822e-06, "elapsed_s": 288.32046461105347}
+{"event": "train", "step": 11800, "loss": 3.0213152885437013, "learning_rate": 9.473492720523113e-06, "elapsed_s": 288.7592487335205}
+{"event": "train", "step": 11820, "loss": 3.028946352005005, "learning_rate": 9.134228147959778e-06, "elapsed_s": 289.20521211624146}
+{"event": "train", "step": 11840, "loss": 3.0105010986328127, "learning_rate": 8.801036683673325e-06, "elapsed_s": 289.64333176612854}
+{"event": "train", "step": 11860, "loss": 3.0454469203948973, "learning_rate": 8.473926728640834e-06, "elapsed_s": 290.08931612968445}
+{"event": "train", "step": 11880, "loss": 3.006731855869293, "learning_rate": 8.152906530502296e-06, "elapsed_s": 290.52924156188965}
+{"event": "train", "step": 11900, "loss": 3.0209120750427245, "learning_rate": 7.837984183352658e-06, "elapsed_s": 290.96688413619995}
+{"event": "train", "step": 11920, "loss": 3.0311039447784425, "learning_rate": 7.529167627537748e-06, "elapsed_s": 291.40438771247864}
+{"event": "train", "step": 11940, "loss": 2.9938814520835875, "learning_rate": 7.2264646494538675e-06, "elapsed_s": 291.85314774513245}
+{"event": "train", "step": 11960, "loss": 3.0245128631591798, "learning_rate": 6.929882881351795e-06, "elapsed_s": 292.2932085990906}
+{"event": "train", "step": 11980, "loss": 3.023373007774353, "learning_rate": 6.6394298011441255e-06, "elapsed_s": 292.7324368953705}
+{"event": "train", "step": 12000, "loss": 3.031375026702881, "learning_rate": 6.355112732216811e-06, "elapsed_s": 293.17010736465454}
+{"event": "validation", "step": 12000, "loss": 4.111901819705963, "recall_at_1": 0.0125}
+{"event": "train", "step": 12020, "loss": 2.9910447716712953, "learning_rate": 6.076938843244423e-06, "elapsed_s": 293.9184465408325}
+{"event": "train", "step": 12040, "loss": 3.014104402065277, "learning_rate": 5.804915148009571e-06, "elapsed_s": 294.3403627872467}
+{"event": "train", "step": 12060, "loss": 2.963191545009613, "learning_rate": 5.53904850522588e-06, "elapsed_s": 294.7947359085083}
+{"event": "train", "step": 12080, "loss": 3.007524609565735, "learning_rate": 5.279345618365128e-06, "elapsed_s": 295.2359199523926}
+{"event": "train", "step": 12100, "loss": 3.016493248939514, "learning_rate": 5.025813035488242e-06, "elapsed_s": 295.6793611049652}
+{"event": "train", "step": 12120, "loss": 3.0018250942230225, "learning_rate": 4.778457149080207e-06, "elapsed_s": 296.1250431537628}
+{"event": "train", "step": 12140, "loss": 3.0224010229110716, "learning_rate": 4.537284195888891e-06, "elapsed_s": 296.55728673934937}
+{"event": "train", "step": 12160, "loss": 3.020830249786377, "learning_rate": 4.302300256767671e-06, "elapsed_s": 297.0371472835541}
+{"event": "train", "step": 12180, "loss": 2.9998428344726564, "learning_rate": 4.073511256522333e-06, "elapsed_s": 297.48987317085266}
+{"event": "train", "step": 12200, "loss": 3.0098670363426208, "learning_rate": 3.850922963761522e-06, "elapsed_s": 297.9240882396698}
+{"event": "train", "step": 12220, "loss": 2.9815412044525145, "learning_rate": 3.6345409907512783e-06, "elapsed_s": 298.3588228225708}
+{"event": "train", "step": 12240, "loss": 3.012010359764099, "learning_rate": 3.424370793273679e-06, "elapsed_s": 298.80247950553894}
+{"event": "train", "step": 12260, "loss": 3.028126609325409, "learning_rate": 3.220417670489112e-06, "elapsed_s": 299.23685026168823}
+{"event": "train", "step": 12280, "loss": 2.981575918197632, "learning_rate": 3.022686764802801e-06, "elapsed_s": 299.6727147102356}
+{"event": "train", "step": 12300, "loss": 3.010413610935211, "learning_rate": 2.831183061735049e-06, "elapsed_s": 300.1098165512085}
+{"event": "train", "step": 12320, "loss": 2.9964461088180543, "learning_rate": 2.6459113897956135e-06, "elapsed_s": 300.5529046058655}
+{"event": "train", "step": 12340, "loss": 2.993246066570282, "learning_rate": 2.466876420361863e-06, "elapsed_s": 300.9871838092804}
+{"event": "train", "step": 12360, "loss": 3.0408135414123536, "learning_rate": 2.294082667561148e-06, "elapsed_s": 301.53401041030884}
+{"event": "train", "step": 12380, "loss": 3.00829439163208, "learning_rate": 2.127534488156807e-06, "elapsed_s": 301.97566986083984}
+{"event": "train", "step": 12400, "loss": 3.0096137523651123, "learning_rate": 1.967236081438395e-06, "elapsed_s": 302.41862988471985}
+{"event": "train", "step": 12420, "loss": 2.9713577389717103, "learning_rate": 1.8131914891158786e-06, "elapsed_s": 302.85665917396545}
+{"event": "train", "step": 12440, "loss": 3.0114706158638, "learning_rate": 1.6654045952175789e-06, "elapsed_s": 303.3035309314728}
+{"event": "train", "step": 12460, "loss": 3.019148898124695, "learning_rate": 1.523879125992389e-06, "elapsed_s": 303.7384150028229}
+{"event": "train", "step": 12480, "loss": 3.0084977984428405, "learning_rate": 1.3886186498157094e-06, "elapsed_s": 304.1915771961212}
+{"event": "train", "step": 12500, "loss": 3.0041406989097594, "learning_rate": 1.2596265770995208e-06, "elapsed_s": 304.6275312900543}
+{"event": "train", "step": 12520, "loss": 2.9706676721572878, "learning_rate": 1.1369061602064524e-06, "elapsed_s": 305.0652551651001}
+{"event": "train", "step": 12540, "loss": 3.034577560424805, "learning_rate": 1.0204604933676531e-06, "elapsed_s": 305.5026421546936}
+{"event": "train", "step": 12560, "loss": 2.953182601928711, "learning_rate": 9.102925126048267e-07, "elapsed_s": 305.9488580226898}
+{"event": "train", "step": 12580, "loss": 2.9509390234947204, "learning_rate": 8.06404995656318e-07, "elapsed_s": 306.39634466171265}
+{"event": "train", "step": 12600, "loss": 2.998723363876343, "learning_rate": 7.088005619068916e-07, "elapsed_s": 306.8380060195923}
+{"event": "train", "step": 12620, "loss": 2.997272717952728, "learning_rate": 6.174816723217847e-07, "elapsed_s": 307.2799699306488}
+{"event": "train", "step": 12640, "loss": 2.9779765486717222, "learning_rate": 5.324506293846732e-07, "elapsed_s": 307.72904872894287}
+{"event": "train", "step": 12660, "loss": 2.994748020172119, "learning_rate": 4.537095770395516e-07, "elapsed_s": 308.1812524795532}
+{"event": "train", "step": 12680, "loss": 3.026642906665802, "learning_rate": 3.8126050063680416e-07, "elapsed_s": 308.61969780921936}
+{"event": "train", "step": 12700, "loss": 2.982159769535065, "learning_rate": 3.151052268829113e-07, "elapsed_s": 309.0561134815216}
+{"event": "train", "step": 12720, "loss": 2.977756977081299, "learning_rate": 2.552454237946533e-07, "elapsed_s": 309.50024604797363}
+{"event": "train", "step": 12740, "loss": 2.9979318380355835, "learning_rate": 2.0168260065689393e-07, "elapsed_s": 309.94010281562805}
+{"event": "train", "step": 12760, "loss": 3.030204784870148, "learning_rate": 1.54418107984583e-07, "elapsed_s": 310.3889322280884}
+{"event": "train", "step": 12780, "loss": 2.9838173508644106, "learning_rate": 1.1345313748870045e-07, "elapsed_s": 310.8308804035187}
+{"event": "train", "step": 12800, "loss": 3.016371154785156, "learning_rate": 7.878872204616916e-08, "elapsed_s": 311.26987528800964}
+{"event": "train", "step": 12820, "loss": 3.000972270965576, "learning_rate": 5.0425735673848004e-08, "elapsed_s": 311.708135843277}
+{"event": "train", "step": 12840, "loss": 3.0257404446601868, "learning_rate": 2.836489350654947e-08, "elapsed_s": 312.1482746601105}
+{"event": "train", "step": 12860, "loss": 3.0216777443885805, "learning_rate": 1.2606751778859727e-08, "elapsed_s": 312.6070246696472}
+{"event": "train", "step": 12880, "loss": 3.022484469413757, "learning_rate": 3.1517078112330844e-09, "elapsed_s": 313.06347823143005}
+{"event": "train", "step": 12900, "loss": 2.997900354862213, "learning_rate": 0.0, "elapsed_s": 313.49029421806335}
+{"event": "final", "step": 12900, "train_loss": 2.999325752258301, "val_loss": 4.116906225681305, "val_recall_at_1": 0.0115}

paper_experiments/2026-09-13/release.json

@@ -1,13 +1,19 @@
-{
-  "release_date": "2026-09-13",
-  "repository": "https://github.com/crimsonKn1ght/galaxy-image-spectrum-clip",
-  "git_commit": "bda764c273e6663350a1d81a08bdedde119b0714",
-  "feature_manifest_sha256": "56336b9ebf6e61aafe8847c54ef9d6cd099d7bf78e0de9dbebc64caa8f85ef15",
-  "gpu_evidence_sha256": "0abfa718a0e617a197505647cb04e0b36f2db98fb713d90fe40bdba1bb50d89e",
-  "cpu_evidence_sha256": "4400b234702b60e36bf690414a42882c383faac96f83fdf6e509c58ede415118",
-  "checkpoint_upload_status": {
-    "seed_42": "already present at repository root",
-    "seed_43": "pending persistent-volume attachment",
-    "seed_44": "pending persistent-volume attachment"
-  }
+{
+  "release_date": "2026-09-13",
+  "repository": "https://github.com/crimsonKn1ght/galaxy-image-spectrum-clip",
+  "git_commit": "bda764c273e6663350a1d81a08bdedde119b0714",
+  "feature_manifest_sha256": "56336b9ebf6e61aafe8847c54ef9d6cd099d7bf78e0de9dbebc64caa8f85ef15",
+  "gpu_evidence_sha256": "0abfa718a0e617a197505647cb04e0b36f2db98fb713d90fe40bdba1bb50d89e",
+  "cpu_evidence_sha256": "4400b234702b60e36bf690414a42882c383faac96f83fdf6e509c58ede415118",
+  "checkpoint_upload_status": {
+    "seed_42": "already present at repository root",
+    "seed_43": "best and final uploaded and verified",
+    "seed_44": "best and final uploaded and verified"
+  },
+  "checkpoint_sha256": {
+    "checkpoints/seed43/best/model.safetensors": "5951b5617447563ea514ebce215b946015f7e9d1315eb5c26db0712cec2077a6",
+    "checkpoints/seed43/final/model.safetensors": "07a6ced8c72535cd20240f0c7d04f09e131ef9447d6b0604b370e515f337073e",
+    "checkpoints/seed44/best/model.safetensors": "d5b20bc048de14602a91a3ee8730ef0b059a2a091083b5bd631fef9a95b39fdf",
+    "checkpoints/seed44/final/model.safetensors": "f81347a7cfb6694850d461140d58cd586d94d66e585c2836b08b9b9fb4009f66"
+  }
 }