{
    "_name_or_path": "/covalent/.cache/models/tmp3kfv_gz97cby61cu",
    "alibi": false,
    "apply_residual_connection_post_layernorm": false,
    "architectures": [
        "FalconForCausalLM"
    ],
    "attention_dropout": 0.0,
    "auto_map": {
        "AutoConfig": "configuration_falcon.FalconConfig",
        "AutoModel": "tiiuae/falcon-7b--modeling_falcon.FalconModel",
        "AutoModelForCausalLM": "tiiuae/falcon-7b--modeling_falcon.FalconForCausalLM",
        "AutoModelForQuestionAnswering": "tiiuae/falcon-7b--modeling_falcon.FalconForQuestionAnswering",
        "AutoModelForSequenceClassification": "tiiuae/falcon-7b--modeling_falcon.FalconForSequenceClassification",
        "AutoModelForTokenClassification": "tiiuae/falcon-7b--modeling_falcon.FalconForTokenClassification"
    },
    "bias": false,
    "bos_token_id": 65024,
    "eos_token_id": 65025,
    "hidden_dropout": 0.0,
    "hidden_size": 4544,
    "initializer_range": 0.02,
    "layer_norm_epsilon": 1e-05,
    "model_type": "falcon",
    "multi_query": true,
    "new_decoder_architecture": false,
    "num_attention_heads": 71,
    "num_hidden_layers": 32,
    "num_kv_heads": 71,
    "pad_token_id": 65025,
    "parallel_attn": true,
    "quantization_config": {
        "_load_in_4bit": false,
        "_load_in_8bit": true,
        "bnb_4bit_compute_dtype": "bfloat16",
        "bnb_4bit_quant_storage": "uint8",
        "bnb_4bit_quant_type": "fp4",
        "bnb_4bit_use_double_quant": false,
        "llm_int8_enable_fp32_cpu_offload": false,
        "llm_int8_has_fp16_weight": false,
        "llm_int8_skip_modules": [
            "lm_head"
        ],
        "llm_int8_threshold": 6.0,
        "load_in_4bit": false,
        "load_in_8bit": true,
        "quant_method": "bitsandbytes"
    },
    "torch_dtype": "float16",
    "transformers_version": "4.46.2",
    "use_cache": true,
    "vocab_size": 65026,
    "api_key": null
}