Update config.json

config.json

@@ -1,7 +1,7 @@
 {
   "architectures": ["GptOssForCausalLM"],
   "model_type": "gpt_oss",
-
+  
   "hidden_size": 2880,
   "intermediate_size": 2880,
   "num_hidden_layers": 24,
@@ -12,17 +12,17 @@
   "head_dim": 64,
   "hidden_act": "silu",
   "initializer_range": 0.02,
-
+  
   "experts_per_token": 4,
   "num_experts_per_tok": 4,
   "num_local_experts": 32,
   "output_router_logits": false,
   "router_aux_loss_coef": 0.9,
-
+  
   "max_position_embeddings": 131072,
   "initial_context_length": 4096,
   "sliding_window": 128,
-
+  
   "rope_scaling": {
     "beta_fast": 32.0,
     "beta_slow": 1.0,
@@ -32,7 +32,7 @@
     "truncate": false
   },
   "rope_theta": 150000,
-
+  
   "layer_types": [
     "sliding_attention", "full_attention",
     "sliding_attention", "full_attention",
@@ -47,19 +47,19 @@
     "sliding_attention", "full_attention",
     "sliding_attention", "full_attention"
   ],
-
+  
   "vocab_size": 201088,
   "eos_token_id": 200002,
   "pad_token_id": 199999,
-
+  
   "rms_norm_eps": 1e-05,
   "swiglu_limit": 7.0,
   "tie_word_embeddings": false,
   "use_cache": true,
-
+  
   "torch_dtype": "float16",
   "transformers_version": "4.55.0",
-
+  
   "quantization_config": {
     "_load_in_4bit": true,
     "_load_in_8bit": false,
@@ -74,11 +74,5 @@
     "llm_int8_threshold": 6.0,
     "llm_int8_skip_modules": ["router", "lm_head", "embed_tokens"],
     "quant_method": "bitsandbytes"
-  },
-
-  "auto_map": {
-    "AutoConfig": "configuration_gpt_oss.GptOssConfig",
-    "AutoModelForCausalLM": "modeling_gpt_oss.GptOssForCausalLM",
-    "AutoTokenizer": "tokenization_gpt_oss.GptOssTokenizer"
   }
-}
+}