Update config.json

config.json

@@ -1,29 +1,43 @@
 {
   "_name_or_path": "Meforgers/Aixr",
   "architectures": [
-    "AixrAI"
+    "Qwen2ForCausalLM"
   ],
   "attention_dropout": 0.0,
   "bos_token_id": 151643,
   "eos_token_id": 151645,
   "hidden_act": "silu",
-  "hidden_size": 3584,
+  "hidden_size": 8192,
   "initializer_range": 0.02,
-  "intermediate_size": 18944,
+  "intermediate_size": 29568,
   "max_position_embeddings": 32768,
-  "max_window_layers": 28,
-  "model_type": "aixr",
-  "num_attention_heads": 28,
-  "num_hidden_layers": 28,
-  "num_key_value_heads": 4,
+  "max_window_layers": 80,
+  "model_type": "qwen2",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": true,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
   "rms_norm_eps": 1e-06,
   "rope_theta": 1000000.0,
   "sliding_window": 131072,
   "tie_word_embeddings": false,
-  "torch_dtype": "float16",
+  "torch_dtype": "bfloat16",
   "transformers_version": "4.41.2",
-  "unsloth_version": "2024.5",
   "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 152064
-}
+}