mlc-chat-config.json

{
  "version": "0.1.0",
  "model_type": "llama",
  "quantization": "q4f16_ft",
  "model_config": {
    "hidden_size": 4096,
    "intermediate_size": 11008,
    "num_attention_heads": 32,
    "num_hidden_layers": 32,
    "rms_norm_eps": 1e-05,
    "vocab_size": 32000,
    "tie_word_embeddings": false,
    "position_embedding_base": 10000,
    "rope_scaling": null,
    "context_window_size": 4096,
    "prefill_chunk_size": 4096,
    "num_key_value_heads": 32,
    "head_dim": 128,
    "tensor_parallel_shards": 1,
    "pipeline_parallel_stages": 1,
    "max_batch_size": 128
  },
  "vocab_size": 32000,
  "context_window_size": 4096,
  "sliding_window_size": -1,
  "prefill_chunk_size": 4096,
  "attention_sink_size": -1,
  "tensor_parallel_shards": 1,
  "pipeline_parallel_stages": 1,
  "temperature": 0.6,
  "presence_penalty": 0.0,
  "frequency_penalty": 0.0,
  "repetition_penalty": 1.0,
  "top_p": 0.9,
  "tokenizer_files": [
    "tokenizer.model",
    "tokenizer.json",
    "tokenizer_config.json"
  ],
  "tokenizer_info": {
    "token_postproc_method": "byte_fallback",
    "prepend_space_in_encode": true,
    "strip_space_in_decode": true
  },
  "conv_template": {
    "name": "llama-2",
    "system_template": "[INST] <<SYS>>\n{system_message}\n<</SYS>>\n\n",
    "system_message": "You are a helpful, respectful and honest assistant.",
    "system_prefix_token_ids": [
      1
    ],
    "add_role_after_system_message": false,
    "roles": {
      "user": "<s>[INST]",
      "assistant": "[/INST]",
      "tool": "[INST]"
    },
    "role_templates": {
      "user": "{user_message}",
      "assistant": "{assistant_message}",
      "tool": "{tool_message}"
    },
    "messages": [],
    "seps": [
      " ",
      " </s>"
    ],
    "role_content_sep": " ",
    "role_empty_sep": " ",
    "stop_str": [
      "[INST]"
    ],
    "stop_token_ids": [
      2
    ],
    "function_string": "",
    "use_function_calling": false
  },
  "pad_token_id": 0,
  "bos_token_id": 1,
  "eos_token_id": 2
}