mlc-chat-config.json

{
  "version": "0.1.0",
  "model_type": "llama",
  "quantization": "q4f32_1",
  "model_config": {
    "hidden_size": 4096,
    "intermediate_size": 14336,
    "num_attention_heads": 32,
    "num_hidden_layers": 32,
    "rms_norm_eps": 1e-05,
    "vocab_size": 128256,
    "position_embedding_base": 500000.0,
    "context_window_size": 8192,
    "prefill_chunk_size": 8192,
    "num_key_value_heads": 8,
    "head_dim": 128,
    "tensor_parallel_shards": 1,
    "max_batch_size": 128
  },
  "vocab_size": 128256,
  "context_window_size": 8192,
  "sliding_window_size": -1,
  "prefill_chunk_size": 8192,
  "attention_sink_size": -1,
  "tensor_parallel_shards": 1,
  "temperature": 0.6,
  "presence_penalty": 0.0,
  "frequency_penalty": 0.0,
  "repetition_penalty": 1.0,
  "top_p": 0.9,
  "tokenizer_files": [
    "tokenizer.json",
    "tokenizer_config.json"
  ],
  "tokenizer_info": {
    "token_postproc_method": "byte_level",
    "prepend_space_in_encode": false,
    "strip_space_in_decode": false
  },
  "conv_template": {
    "name": "llama-3",
    "system_template": "<|start_header_id|>system<|end_header_id|>\n\n{system_message}<|eot_id|>\n",
    "system_message": "You are a helpful, respectful and honest assistant.",
    "system_prefix_token_ids": [
      128000
    ],
    "add_role_after_system_message": true,
    "roles": {
      "user": "<|start_header_id|>user",
      "assistant": "<|start_header_id|>assistant"
    },
    "role_templates": {
      "user": "{user_message}",
      "assistant": "{assistant_message}",
      "tool": "{tool_message}"
    },
    "messages": [],
    "seps": [
      "<|eot_id|>"
    ],
    "role_content_sep": "<|end_header_id|>\n\n",
    "role_empty_sep": "<|end_header_id|>\n\n",
    "stop_str": [
      "<|end_of_text|>",
      "<|eot_id|>"
    ],
    "stop_token_ids": [
      128001,
      128009
    ],
    "function_string": "",
    "use_function_calling": false
  },
  "pad_token_id": 0,
  "bos_token_id": 128000,
  "eos_token_id": [
    128001,
    128009
  ],
  "mean_gen_len": 128,
  "max_gen_len": 512,
  "shift_fill_factor": 0.3
}