Illustrating Reinforcement Learning from Human Feedback (RLHF)
•
119
Alex Havrilla
Dahoas
AI & ML interests
NLP, RL
Recent Activity
updated
a dataset
2 days ago
Dahoas/numina-synthetic
updated
a dataset
14 days ago
Dahoas/aimo-validation-aime
Articles
Organizations
models
33
Dahoas/gptj-rm-IHP
Updated
•
2
Dahoas/gptneox-response-full-static-sft
Text Generation
•
Updated
•
13
•
1
Dahoas/pythia-1B-response-full-static-sft
Text Generation
•
Updated
•
17
•
1
Dahoas/pythia-125M-response-full-static-sft
Text Generation
•
Updated
•
13
•
1
Dahoas/synthetic-pythia-6B-rm-sft-response
Text Generation
•
Updated
•
17
Dahoas/pythia-6B-sft-response-full-static
Text Generation
•
Updated
•
18
•
1
Dahoas/gptj-6B-response-full-static-sft
Text Generation
•
Updated
•
14
•
1
Dahoas/pythia-6B-rm-response-full-hh
Updated
Dahoas/gptj-response-full-sft
Text Generation
•
Updated
•
12
•
1
Dahoas/pythia-6b-rm-response-only-full-hh
Text Generation
•
Updated
•
14
datasets
147
Dahoas/numina-synthetic
Viewer
•
Updated
•
361k
•
117
Dahoas/aimo-validation-aime
Viewer
•
Updated
•
90
•
15
Dahoas/qwen-1.5-4B-default-positives-epoch-1-100
Viewer
•
Updated
•
290k
•
43
Dahoas/qwen-1.5-4B-tree-positives-epoch-2-100
Viewer
•
Updated
•
491k
•
41
Dahoas/qwen-1.5-4B-tree-positives-epoch-1-100
Viewer
•
Updated
•
477k
•
46
Dahoas/qwen-1.5-4B-epoch-1-test-100
Viewer
•
Updated
•
498k
•
62
Dahoas/qwen-1.5-4B-K-100-test
Viewer
•
Updated
•
500k
•
31
Dahoas/MATH_train_K_100_qwen_1.5_4B_outputs
Viewer
•
Updated
•
750k
•
36
Dahoas/MATH-K-100-train
Viewer
•
Updated
•
750k
•
1.25k
•
2
Dahoas/gsm8k_reformatted
Viewer
•
Updated
•
8.79k
•
37