deepseek-ai/deepseek-coder-7b-base-v1.5/bert4torch_config.json

{
  "model": "llama",
  "template": "pretrained_text_continuation",
  "bos_token_id": 100000,
  "eos_token_id": 100015,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 4096,
  "num_attention_heads": 32,
  "num_hidden_layers": 30,
  "num_key_value_heads": 32,
  "layer_norm_eps": 1e-06,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "vocab_size": 102400,
  "rope_rank": "updown",
  "_attn_implementation": "sdpa",
  "segment_vocab_size": 0,
  "skip_init": true,
  "convert_lm_logits_dtype": "float32",
  "generation_config": {"tokenizer_config":  {"skip_special_tokens": true, "add_special_tokens": false}, "eos_token_id": [100015]}
}