config.json

{
  "architectures": [
    "LlamaForCausalLMEagle3"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "draft_vocab_size": 32000,
  "dtype": "bfloat16",
  "eagle_config": {
    "eagle_aux_hidden_state_layer_ids": [
      2,
      12,
      20
    ]
  },
  "eos_token_id": 2,
  "fc_norm": true,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2880,
  "initializer_range": 0.02,
  "intermediate_size": 23040,
  "max_position_embeddings": 131072,
  "mlp_bias": false,
  "model_type": "llama",
  "norm_output": true,
  "num_attention_heads": 64,
  "num_hidden_layers": 1,
  "num_key_value_heads": 16,
  "pad_token_id": 0,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_parameters": {
    "beta_fast": 32.0,
    "beta_slow": 1.0,
    "factor": 32.0,
    "mscale": 1,
    "mscale_all_dim": 0,
    "original_max_position_embeddings": 4096,
    "rope_theta": 150000,
    "rope_type": "yarn",
    "truncate": false
  },
  "tie_word_embeddings": false,
  "transformers_version": "5.3.0",
  "use_cache": true,
  "vocab_size": 201088
}