config.json

{
  "architectures": [
    "MoonshotForCausalLM"
  ],
  "attention_bias": false,
  "auto_map": {
    "AutoConfig": "configuration_moonshot.MoonshotConfig",
    "AutoModel": "modeling_moonshot.MoonshotModel",
    "AutoModelForCausalLM": "modeling_moonshot.MoonshotForCausalLM"
  },
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 3072,
  "initializer_range": 0.02,
  "intermediate_size": 8192,
  "model_type": "moonshot",
  "num_attention_heads": 24,
  "num_hidden_layers": 16,
  "num_key_value_heads": 8,
  "pad_token_id": 0,
  "pretraining_sequence_length": 40000,
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "factor": 16.0,
    "type": "linear"
  },
  "rope_theta": 500000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.46.1",
  "use_cache": true,
  "vocab_size": 172032
}