上传 config.json

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/aifs4su/data/zheny/speechllm/logs_llasa/data_v12_25_tts_25w_mix_final_1b/checkpoint-240000",
+  "_name_or_path": "/aifs4su/data/zheny/speechllm/logs_llasa/data_v1_2_tts_25w_mix_final_8b/checkpoint-125000",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -11,28 +11,28 @@
     128008,
     128009
   ],
-  "head_dim": 64,
+  "head_dim": 128,
   "hidden_act": "silu",
-  "hidden_size": 2048,
+  "hidden_size": 4096,
   "initializer_range": 0.02,
-  "intermediate_size": 8192,
+  "intermediate_size": 14336,
   "max_position_embeddings": 131072,
   "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 32,
-  "num_hidden_layers": 16,
+  "num_hidden_layers": 32,
   "num_key_value_heads": 8,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": {
-    "factor": 32.0,
+    "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": true,
+  "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.46.1",
   "use_cache": true,