Training loss: 2.5767

config.json

@@ -7,23 +7,23 @@
   "bos_token_id": 0,
   "eos_token_id": 2,
   "hidden_act": "silu",
-  "hidden_size": 660,
+  "hidden_size": 576,
   "initializer_range": 0.02,
-  "intermediate_size": 2640,
-  "max_position_embeddings": 128,
+  "intermediate_size": 2304,
+  "max_position_embeddings": 512,
   "mlp_bias": false,
   "model_type": "llama",
-  "num_attention_heads": 10,
-  "num_hidden_layers": 20,
-  "num_key_value_heads": 10,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "num_key_value_heads": 12,
   "pad_token_id": 1,
   "pretraining_tp": 1,
-  "rms_norm_eps": 1e-06,
+  "rms_norm_eps": 1e-05,
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.44.2",
   "use_cache": true,
-  "vocab_size": 52000
+  "vocab_size": 32000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1abece3f8701f18960454fb1414238177c0a00d2b90eb8db16004b8b6c991b78
-size 832256624
+oid sha256:0c5a8800275bc25d8caa3d29a7f5e2eb721cf02074834584fadea457b0ad9173
+size 402329888