Rename chess-mamba-vs-xformer/config/Mamba/50m.py to chess-mamba-vs-xformer/config/Mamba/50M.py

Delete chess-mamba-vs-xformer/config/Mamba/50M.py

Upload chess-mamba-vs-xformer/config/Mamba/50m.py with huggingface_hub

Upload 450 files

Rename chess-mamba-vs-xformer/out/Mamba/50M/anneal/old_anneal_complete.pt to chess-mamba-vs-xformer/out/Mamba/50M/anneal/old_v2_anneal_complete.pt

Rename chess-mamba-vs-xformer/out/Mamba/50M/ckpt_35723241.pt to chess-mamba-vs-xformer/out/Mamba/50M/old_v2_ckpt_35723241.pt

Rename chess-mamba-vs-xformer/out/Mamba/50M/ckpt_22275000.pt.pt to chess-mamba-vs-xformer/out/Mamba/50M/old_v2_ckpt_22275000.pt

Rename chess-mamba-vs-xformer/out/Mamba/50M/ckpt_12653100.pt to chess-mamba-vs-xformer/out/Mamba/50M/old_v1_ckpt_12653100.pt

Rename chess-mamba-vs-xformer/out/Mamba/50M/anneal/anneal_complete.pt to chess-mamba-vs-xformer/out/Mamba/50M/anneal/old_anneal_complete.pt

Rename chess-mamba-vs-xformer/out/Xformer/50M/anneal/anneal_complete_wrongds.pt to chess-mamba-vs-xformer/out/Xformer/50M/anneal/old_anneal_complete_wrongds.pt

Rename chess-mamba-vs-xformer/out/Xformer/50M/anneal/anneal_complete.pt to chess-mamba-vs-xformer/out/Xformer/50M/anneal/old_anneal_complete.pt

Update chess-mamba-vs-xformer/config/Mamba/50M.py

Update README.md

Delete chess-mamba-vs-xformer/data

Create data/

Delete chess-mamba-vs-xformer/data/anneal/anneal.zip

Delete chess-mamba-vs-xformer/data/stable/stable.zip

Update chess-mamba-vs-xformer/config/Mamba/50M.py

Upload chess-gpt-eval/logs/50M/ckpt_35723241_vs_stockfish_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp/exp_P2B_10_25M(annealed)_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp/exp_P2A_10_25M(annealed)_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp//exp_P1_2C_3_03M_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp//exp_P1_2B_3_03M_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp//exp_P1_1B_3_03M_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/50M/ckpt_25457100_vs_stockfish_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp//exp_P1_1A_3_03M_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/exp/exp_P1_2A_vs_lc0_sweep.csv with huggingface_hub

Delete chess-gpt-eval/logs/exp

Upload chess-gpt-eval/logs/50M/ckpt_22275000_pt_vs_stockfish_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/50M/ckpt_22275000_pt_vs_stockfish_sweep.csv with huggingface_hub

Delete chess-gpt-eval/logs/50M/anneal/anneal_complete_pt_vs_stockfish_sweep.csv

Delete chess-gpt-eval/logs/50M/anneal/anneal_complete_pt_vs_lc0_sweep.csv

Delete chess-gpt-eval/logs/50M/ckpt_35723241_pt_vs_lc0_sweep.csv

Rename chess-gpt-eval/logs/50M/ckpt_12653100_pt_vs_lc0_sweep.csv to chess-gpt-eval/logs/50M/ckpt_12653100_pt_vs_stockfish_sweep.csv

Delete chess-gpt-eval/logs/50M/ckpt_22275000_pt_vs_stockfish_sweep.csv

Upload chess-gpt-eval/logs/50M/ckpt_12653100_pt_vs_lc0_sweep.csv with huggingface_hub

Delete csv_datasets

Fixed early-stopping in get_mamba_response based on space/dot tokens (now decodes the strings instead of using hardcoded token ids).

Update chess-gpt-eval/main.py

Delete chess-mamba-vs-xformer/out/Xformer/50M/ckpt_7740000b.pt

Delete chess-mamba-vs-xformer/out/Mamba/50M/ckpt_10052100.pt

Delete chess-mamba-vs-xformer/out/Mamba/50M/ckpt_11704500.pt

Delete chess-mamba-vs-xformer/out/Mamba/50M/ckpt_7512300.pt

Upload chess-gpt-eval/logs/11M_ds16_lr5_dmodel288_20layer/ckpt_867000_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/11M_ds16_lr5_dmodel288_20layer/ckpt_1054000_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/11M_ds16_lr5/ckpt_2550000_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/11M_ds16_lr5/ckpt_1054000_pt_vs_lc0_sweep.csv with huggingface_hub

Delete chess-gpt-eval/logs/11M_ds16_lr5/ckpt_1054000_pt_vs_lc0_sweep.csv

Upload chess-gpt-eval/logs/11M_ds16_lr5/ckpt_1054000_pt_vs_lc0_sweep.csv with huggingface_hub

Upload chess-gpt-eval/logs/11M_ds16/ckpt_6069000_pt_vs_lc0_sweep.csv with huggingface_hub