{
  "_name_or_path": "BAAI/bge-reranker-v2.5-gemma2-lightweight",
  "architectures": [
    "CostWiseGemmaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attn_logit_softcapping": 50.0,
  "auto_map": {
    "AutoConfig": "BAAI/bge-reranker-v2.5-gemma2-lightweight--gemma_config.CostWiseGemmaConfig",
    "AutoModel": "BAAI/bge-reranker-v2.5-gemma2-lightweight--gemma_model.CostWiseGemmaModel",
    "AutoModelForCausalLM": "BAAI/bge-reranker-v2.5-gemma2-lightweight--gemma_model.CostWiseGemmaForCausalLM"
  },
  "bos_token_id": 2,
  "cache_implementation": "hybrid",
  "eos_token_id": 1,
  "final_logit_softcapping": 30.0,
  "head_dim": 256,
  "hidden_act": "gelu_pytorch_tanh",
  "hidden_activation": "gelu_pytorch_tanh",
  "hidden_size": 3584,
  "initializer_range": 0.02,
  "intermediate_size": 14336,
  "layer_sep": 1,
  "layer_wise": true,
  "max_position_embeddings": 8192,
  "model_type": "cost_wise_gemma",
  "num_attention_heads": 16,
  "num_hidden_layers": 42,
  "num_key_value_heads": 8,
  "pad_token_id": 0,
  "query_pre_attn_scalar": 256,
  "rms_norm_eps": 1e-06,
  "rope_theta": 10000.0,
  "sliding_window": 4096,
  "sliding_window_size": 4096,
  "start_layer": 8,
  "torch_dtype": "float16",
  "transformers_version": "4.43.4",
  "use_cache": true,
  "vocab_size": 256000
}