Fix latest config to flat LlamaConfig format

latest/config.json

@@ -1,69 +1,31 @@
 {
   "architectures": [
-    "Eagle3Speculator"
+    "LlamaForCausalLM"
   ],
-  "auto_map": {
-    "": "config.PEagleSpeculatorConfig"
-  },
-  "base_model_ep_plan": null,
-  "down_sample_ratio": 0.7,
-  "down_sample_ratio_min": 0.2,
+  "model_type": "llama",
+  "hidden_size": 3072,
+  "intermediate_size": 8192,
+  "num_hidden_layers": 1,
+  "num_attention_heads": 24,
+  "num_key_value_heads": 8,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "max_position_embeddings": 196608,
+  "initializer_range": 0.02,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "vocab_size": 200064,
   "draft_vocab_size": 200064,
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "mlp_bias": false,
+  "tie_word_embeddings": false,
+  "use_cache": true,
   "dtype": "bfloat16",
+  "ptd_token_id": 200063,
   "eagle_aux_hidden_state_layer_ids": [
     2,
     31,
     59
-  ],
-  "embed_requires_grad": true,
-  "has_no_defaults_at_init": false,
-  "max_seq_len": 4096,
-  "norm_before_fc": false,
-  "norm_before_residual": false,
-  "para_depths": 8,
-  "prediction_loss_weight": 1.0,
-  "ptd_token_id": 200063,
-  "speculators_config": {
-    "algorithm": "peagle",
-    "default_proposal_method": "greedy",
-    "proposal_methods": [
-      {
-        "accept_tolerance": 0.0,
-        "proposal_type": "greedy",
-        "speculative_tokens": 8,
-        "verifier_accept_k": 1
-      }
-    ],
-    "verifier": {
-      "architectures": [],
-      "name_or_path": "MiniMaxAI/MiniMax-M2.7"
-    }
-  },
-  "speculators_model_type": "eagle3",
-  "speculators_version": "0.4.0.dev0",
-  "target_hidden_size": null,
-  "tie_word_embeddings": false,
-  "transformer_layer_config": {
-    "attention_bias": false,
-    "attention_dropout": 0.0,
-    "head_dim": 128,
-    "hidden_act": "silu",
-    "hidden_size": 3072,
-    "initializer_range": 0.02,
-    "intermediate_size": 8192,
-    "max_position_embeddings": 196608,
-    "mlp_bias": false,
-    "model_type": "llama",
-    "num_attention_heads": 24,
-    "num_hidden_layers": 1,
-    "num_key_value_heads": 8,
-    "pretraining_tp": 1,
-    "rms_norm_eps": 1e-06,
-    "rope_scaling": null,
-    "rope_theta": 10000.0,
-    "use_cache": true,
-    "vocab_size": 200064,
-    "ptd_token_id": 200063
-  },
-  "transformers_version": "4.57.6"
+  ]
 }