Dahoas/1B_hh_sft_prompted_comparison
Viewer • Updated • 100 • 3
Alex Havrilla
Dahoas
AI & ML interests
NLP, RL
Organizations
Dahoas/125M_hh_sft_prompted_comparison
Viewer • Updated • 100 • 3 • 1
Dahoas/6B_hh_sft_ppo_comparison
Viewer • Updated • 100 • 2
Dahoas/1B_hh_sft_ppo_comparison
Viewer • Updated • 100 • 7
Dahoas/125M_hh_sft_ppo_comparison
Viewer • Updated • 100 • 3 • 1
Dahoas/pythia_6B_prompted_hh_eval_human
Viewer • Updated • 100 • 3
Dahoas/pythia_1B_prompted_hh_eval_human
Viewer • Updated • 100 • 3
Dahoas/pythia_125M_prompted_hh_eval_human
Viewer • Updated • 100 • 3
Dahoas/openai_summarize_tldr_human_eval
Viewer • Updated • 100 • 4 • 1
Dahoas/hh_synthetic_eval
Viewer • Updated • 100 • 3
Dahoas/synthetic_eval
Viewer • Updated • 100 • 3
Dahoas/20B_hybrid_comparison
Viewer • Updated • 100 • 3
Dahoas/6B_hybrid_comparison
Viewer • Updated • 100 • 3
Dahoas/1B_hybrid_comparison
Viewer • Updated • 100 • 3
Dahoas/125M_hybrid_comparison
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_125M_inference_test_human
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_20B_inference_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_1B_inference_test_human
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_6B_inference_test_human
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_20B_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_125M_hybrid_test_human_human
Viewer • Updated • 100 • 3 • 1
Dahoas/pythia_synthetic_1B_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_synthetic_6B_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_125M_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_1B_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_20B_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_6B_hybrid_test_human_human
Viewer • Updated • 100 • 3
Dahoas/pythia_20B_inference_test_human
Viewer • Updated • 100 • 3
Dahoas/pythia_1B_inference_test_human
Viewer • Updated • 100 • 3
Dahoas/pythia_6B_inference_test_human
Viewer • Updated • 100 • 3