Add training logs and README

config.json

@@ -70,7 +70,7 @@
   "num_attention_heads": 16,
   "num_hidden_layers": 48,
   "num_key_value_heads": 8,
-  "pad_token_id": 0,
+  "pad_token_id": 1,
   "query_pre_attn_scalar": 256,
   "rms_norm_eps": 1e-06,
   "rope_local_base_freq": 10000.0,
@@ -81,6 +81,6 @@
   "rope_theta": 1000000.0,
   "sliding_window": 1024,
   "transformers_version": "4.56.0",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 262208
 }