Update stop token and rope scaling to match llama 3.1

config.json

@@ -6,12 +6,16 @@
   "attention_bias": false,
   "attention_dropout": 0.0,
   "bos_token_id": 128000,
-  "eos_token_id": 128001,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
   "hidden_act": "silu",
   "hidden_size": 8192,
   "initializer_range": 0.02,
   "intermediate_size": 28672,
-  "max_position_embeddings": 8192,
+  "max_position_embeddings": 131072,
   "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 64,
@@ -19,7 +23,13 @@
   "num_key_value_heads": 8,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
-  "rope_scaling": null,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
   "rope_theta": 500000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",