The *RLMT* collection. Coming soon!
Princeton NLP group
princeton-nlp
AI & ML interests
None yet
Recent Activity
upvoted a paper about 1 month ago
VisGym: Diverse, Customizable, Scalable Environments for Multimodal Agents new activity
3 months ago
HuggingFaceTB/FineMath-Llama-3B:Hyperparameters updated
a collection
5 months ago
RLMT Experiments Organizations
SWE-bench is a benchmark for evaluating Language Models and AI Systems on their ability resolve real world GitHub Issues.
This collections contains a list of SimPO and baseline models.
-
princeton-nlp/gemma-2-9b-it-SimPO
Text Generation • 9B • Updated • 717 • • 172 -
princeton-nlp/gemma-2-9b-it-DPO
Text Generation • 9B • Updated • 24 • • 9 -
princeton-nlp/Llama-3-Base-8B-SFT-IPO
Text Generation • 8B • Updated • 20 • • 1 -
princeton-nlp/Llama-3-Base-8B-SFT-DPO
Text Generation • 8B • Updated • 765 •
ProLong is a family of long-context models that are continued trained and supervised fine-tuned from Llama-3-8B, with a maximum context window of 512K
-
princeton-nlp/Llama-3-8B-ProLong-64k-Base
Text Generation • 8B • Updated • 8.13k • • 5 -
princeton-nlp/Llama-3-8B-ProLong-64k-Instruct
Text Generation • 8B • Updated • 8.4k • • 13 -
princeton-nlp/Llama-3-8B-ProLong-512k-Base
8B • Updated • 8.06k • 9 -
princeton-nlp/Llama-3-8B-ProLong-512k-Instruct
8B • Updated • 8.17k • 26
-
princeton-nlp/unsup-simcse-bert-base-uncased
Feature Extraction • Updated • 1.38k • • 5 -
princeton-nlp/unsup-simcse-bert-large-uncased
Feature Extraction • Updated • 33 • 1 -
princeton-nlp/unsup-simcse-roberta-base
Feature Extraction • Updated • 588 • 9 -
princeton-nlp/unsup-simcse-roberta-large
Feature Extraction • Updated • 92 • 3
The *RLMT* collection. Coming soon!
This collections contains a list of SimPO and baseline models.
-
princeton-nlp/gemma-2-9b-it-SimPO
Text Generation • 9B • Updated • 717 • • 172 -
princeton-nlp/gemma-2-9b-it-DPO
Text Generation • 9B • Updated • 24 • • 9 -
princeton-nlp/Llama-3-Base-8B-SFT-IPO
Text Generation • 8B • Updated • 20 • • 1 -
princeton-nlp/Llama-3-Base-8B-SFT-DPO
Text Generation • 8B • Updated • 765 •
SWE-bench is a benchmark for evaluating Language Models and AI Systems on their ability resolve real world GitHub Issues.
ProLong is a family of long-context models that are continued trained and supervised fine-tuned from Llama-3-8B, with a maximum context window of 512K
-
princeton-nlp/Llama-3-8B-ProLong-64k-Base
Text Generation • 8B • Updated • 8.13k • • 5 -
princeton-nlp/Llama-3-8B-ProLong-64k-Instruct
Text Generation • 8B • Updated • 8.4k • • 13 -
princeton-nlp/Llama-3-8B-ProLong-512k-Base
8B • Updated • 8.06k • 9 -
princeton-nlp/Llama-3-8B-ProLong-512k-Instruct
8B • Updated • 8.17k • 26
-
princeton-nlp/unsup-simcse-bert-base-uncased
Feature Extraction • Updated • 1.38k • • 5 -
princeton-nlp/unsup-simcse-bert-large-uncased
Feature Extraction • Updated • 33 • 1 -
princeton-nlp/unsup-simcse-roberta-base
Feature Extraction • Updated • 588 • 9 -
princeton-nlp/unsup-simcse-roberta-large
Feature Extraction • Updated • 92 • 3