Upload config.json with huggingface_hub
Browse files- config.json +23 -0
config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": "ChessTransformer200M",
|
| 3 |
+
"params": 204006404,
|
| 4 |
+
"encoder_dim": 256,
|
| 5 |
+
"hidden_dim": 1024,
|
| 6 |
+
"num_layers": 16,
|
| 7 |
+
"num_heads": 16,
|
| 8 |
+
"ffn_ratio": 4,
|
| 9 |
+
"policy_head_dim": 512,
|
| 10 |
+
"value_hidden": 512,
|
| 11 |
+
"vocab_size": 5504,
|
| 12 |
+
"training": {
|
| 13 |
+
"steps": 10000,
|
| 14 |
+
"positions_seen": 10240256,
|
| 15 |
+
"batch_size": 256,
|
| 16 |
+
"accum_steps": 4,
|
| 17 |
+
"effective_batch": 1024,
|
| 18 |
+
"lr": 0.0002,
|
| 19 |
+
"best_accuracy": 0.184,
|
| 20 |
+
"dataset": "avewright/chess-positions-lichess-sf",
|
| 21 |
+
"gpu": "NVIDIA A40 46GB"
|
| 22 |
+
}
|
| 23 |
+
}
|