Collection of datasets for DPO for development. Data come from clembench v0.9 and v1.0 for all games, except for referencegame (v1.6).
clembench-project-playpen
community
AI & ML interests
None defined yet.
Recent Activity
View all activity
Collections
4
models
306
clembench-playpen/llama-3.1-8B-Instruct-multi-step-collator_playpen_SFT-e3_DFINAL_0.7K-steps
Updated
•
2
clembench-playpen/llama3.1_DPO_2neg_Aborted_best_models_old_taboo_only_DPO_noSFT_______
Updated
clembench-playpen/llama3.1_DPO_2neg_Aborted_best_models_old_exp_DPO_noSFT_______
Updated
clembench-playpen/meta-llama-Meta-Llama-3.1-8B-Instruct_SFT_E1_D40005_Michael_unload_from4bit
Text Generation
•
Updated
clembench-playpen/meta-llama-3.1-8b-instruct-unsloth-bnb-4bit_KTO_Aborted_best_models_F_KTO_noSFT
Updated
clembench-playpen/llama3.1_DPO_2neg_Aborted_best_models_old_LA_DPO_noSFT
Updated
clembench-playpen/llama-3.1-8B-Instruct-only-warmup_playpen_SFT-e3_warm-up_0.225K-steps
Updated
•
31
clembench-playpen/llama-3.1-8B-Instruct-only-warmup_playpen_SFT-e3_warm-up_0.1K-steps
Updated
•
29
clembench-playpen/Mistral-Small-24B-Instruc-0.1k-warmup_playpen_SFT-e3_DFINAL_0.6K-steps
Updated
clembench-playpen/meta-llama-Meta-Llama-3.1-8B-Instruct_SFT_E1_D40005_merged_4bit
Text Generation
•
Updated
•
15
datasets
38
clembench-playpen/DPO_2neg_Aborted_best_models_old_taboo_only
Viewer
•
Updated
•
1.27k
clembench-playpen/DPO_2neg_Aborted_best_models_old_exp
Viewer
•
Updated
•
3.39k
clembench-playpen/warm-up_synthetic-data
Viewer
•
Updated
•
1.35k
•
9
clembench-playpen/DPO_allneg_Aborted_best_models_old_LA
Viewer
•
Updated
•
4.48k
•
36
clembench-playpen/DPO_6neg_Aborted_best_models_old_LA
Viewer
•
Updated
•
4.45k
•
25
clembench-playpen/DPO_5neg_Aborted_best_models_old_LA
Viewer
•
Updated
•
4.38k
•
24
clembench-playpen/DPO_4neg_Aborted_best_models_old_LA
Viewer
•
Updated
•
4.25k
•
23
clembench-playpen/DPO_3neg_Aborted_best_models_old_LA
Viewer
•
Updated
•
3.95k
•
25
clembench-playpen/DPO_2neg_Aborted_old_LA
Viewer
•
Updated
•
12.4k
•
35
clembench-playpen/DPO_2neg_Aborted_same_family_model_old_LA
Viewer
•
Updated
•
6.35k
•
34