avewright commited on
Commit
d74137c
·
verified ·
1 Parent(s): d9e88a9

opening expert model card

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -18,7 +18,7 @@ finetuned on `≥26`-piece positions (one-hot best first move) from
18
  [`Lichess/chess-position-evaluations`](https://huggingface.co/datasets/Lichess/chess-position-evaluations)
19
  via [`avewright/lichess-opening-bestline`](https://huggingface.co/datasets/avewright/lichess-opening-bestline).
20
 
21
- This file is **`latest.pt` at opening-FT step 8000** (2026-09-14 14:41 UTC).
22
  Train loss ~1.7585. Frozen holdout hard CE ~1.4045.
23
 
24
  Not the generalist incumbent, the puzzle expert, the Syzygy expert, or the endgame expert.
 
18
  [`Lichess/chess-position-evaluations`](https://huggingface.co/datasets/Lichess/chess-position-evaluations)
19
  via [`avewright/lichess-opening-bestline`](https://huggingface.co/datasets/avewright/lichess-opening-bestline).
20
 
21
+ This file is **`latest.pt` at opening-FT step 8000** (2026-09-14 14:42 UTC).
22
  Train loss ~1.7585. Frozen holdout hard CE ~1.4045.
23
 
24
  Not the generalist incumbent, the puzzle expert, the Syzygy expert, or the endgame expert.