trainer_state.json

{
  "best_metric": 1.1390750408172607,
  "best_model_checkpoint": "/net/nfs/mosaic/yuchenl/models/llm_blender/fuser_xl_prbar_0529/checkpoint-25",
  "epoch": 0.029994001199760048,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0,
      "loss": 1.0875,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0,
      "loss": 1.1061,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 0.9996,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 1.1832,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.0332,
      "step": 5
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.0285,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.1137,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.1479,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.0779,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.0459,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.0973,
      "step": 11
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.1371,
      "step": 12
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0336,
      "step": 13
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0789,
      "step": 14
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0539,
      "step": 15
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0967,
      "step": 16
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.1219,
      "step": 17
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0523,
      "step": 18
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0473,
      "step": 19
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.0336,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.0443,
      "step": 21
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.1203,
      "step": 22
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.1496,
      "step": 23
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.1021,
      "step": 24
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.0074,
      "step": 25
    },
    {
      "epoch": 0.03,
      "eval_loss": 1.1390750408172607,
      "eval_runtime": 292.2297,
      "eval_samples_per_second": 17.11,
      "eval_steps_per_second": 0.216,
      "step": 25
    }
  ],
  "max_steps": 12495,
  "num_train_epochs": 15,
  "total_flos": 2.2295327973310464e+16,
  "trial_name": null,
  "trial_params": null
}