Chess Challenge submission by Sunxt25

README.md

@@ -14,13 +14,13 @@ Chess model submitted to the LLM Course Chess Challenge.
 ## Submission Info
 
 - **Submitted by**: [Sunxt25](https://huggingface.co/Sunxt25)
-- **Parameters**: 713,600
+- **Parameters**: 921,960
 - **Organization**: LLM-course
 
 ## Model Details
 
 - **Architecture**: Chess Transformer (GPT-style)
-- **Vocab size**: 149
-- **Embedding dim**: 128
-- **Layers**: 4
-- **Heads**: 4
+- **Vocab size**: 153
+- **Embedding dim**: 120
+- **Layers**: 6
+- **Heads**: 8

config.json

@@ -9,12 +9,12 @@
   "layer_norm_epsilon": 1e-05,
   "model_type": "chess_transformer",
   "n_ctx": 256,
-  "n_embd": 128,
-  "n_head": 4,
-  "n_inner": 384,
-  "n_layer": 4,
+  "n_embd": 120,
+  "n_head": 8,
+  "n_inner": 360,
+  "n_layer": 6,
   "pad_token_id": 0,
   "tie_weights": true,
   "transformers_version": "4.57.5",
-  "vocab_size": 149
+  "vocab_size": 153
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:038d7feeb236cc5e7c424d6e65f77bbbe226b4bbb459583b74c22714a01f27d4
-size 2858800
+oid sha256:0a8ba31619215db2c6b9b7c5241f89b5b16e270b72ef1bbac8c37a2cb1445f0b
+size 3694280

vocab.json

@@ -147,5 +147,9 @@
   "(+)": 145,
   "(+*)": 146,
   "(o)": 147,
-  "(O)": 148
+  "(O)": 148,
+  "q": 149,
+  "r": 150,
+  "b": 151,
+  "n": 152
 }