Chess Challenge submission by felixleb

README.md

@@ -13,14 +13,14 @@ Chess model submitted to the LLM Course Chess Challenge.
 
 ## Submission Info
 
-- **Submitted by**: [Eithannak](https://huggingface.co/Eithannak)
-- **Parameters**: 997,872
+- **Submitted by**: [felixleb](https://huggingface.co/felixleb)
+- **Parameters**: 909,824
 - **Organization**: LLM-course
 
 ## Model Details
 
 - **Architecture**: Chess Transformer (GPT-style)
-- **Vocab size**: 5581
-- **Embedding dim**: 96
-- **Layers**: 5
+- **Vocab size**: 1682
+- **Embedding dim**: 128
+- **Layers**: 4
 - **Heads**: 4

config.json

@@ -9,13 +9,12 @@
   "layer_norm_epsilon": 1e-05,
   "model_type": "chess_transformer",
   "n_ctx": 256,
-  "n_embd": 96,
+  "n_embd": 128,
   "n_head": 4,
-  "n_inner": 192,
-  "n_layer": 5,
+  "n_inner": 384,
+  "n_layer": 4,
   "pad_token_id": 0,
-  "rope_theta": 10000.0,
   "tie_weights": true,
-  "transformers_version": "4.57.5",
-  "vocab_size": 5581
+  "transformers_version": "4.57.3",
+  "vocab_size": 1682
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52ecaf7b03785f2be92583d2896a89c8898b29e7b739642bb9a4bc8df515798d
-size 3995496
+oid sha256:afc3a5896b197224752804e7a8f89c3d406972db8958da5af3ac0035bc9a23e3
+size 3643696