mlc-chat-config.json

{
  "model_type": "llama",
  "quantization": "q4f16_1",
  "model_config": {
    "hidden_size": 5120,
    "intermediate_size": 13824,
    "num_attention_heads": 40,
    "num_hidden_layers": 40,
    "rms_norm_eps": 1e-05,
    "vocab_size": 32000,
    "position_embedding_base": 10000,
    "context_window_size": 4096,
    "prefill_chunk_size": 4096,
    "num_key_value_heads": 40,
    "head_dim": 128,
    "tensor_parallel_shards": 1
  },
  "vocab_size": 32000,
  "context_window_size": 4096,
  "sliding_window_size": -1,
  "prefill_chunk_size": 4096,
  "attention_sink_size": -1,
  "tensor_parallel_shards": 1,
  "mean_gen_len": 128,
  "max_gen_len": 512,
  "shift_fill_factor": 0.3,
  "temperature": 0.6,
  "repetition_penalty": 1.0,
  "top_p": 0.9,
  "conv_template": "llama-2",
  "pad_token_id": 0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "tokenizer_files": [
    "tokenizer.model",
    "tokenizer.json",
    "tokenizer_config.json"
  ],
  "version": "0.1.0"
}