config.json

{
    "_remove_final_layer_norm": false,
    "activation_dropout": 0.0,
    "activation_function": "relu",
    "architectures": [
        "OPTForCausalLM"
    ],
    "attention_dropout": 0.0,
    "bos_token_id": 2,
    "do_layer_norm_before": true,
    "dropout": 0.1,
    "dtype": "float16",
    "enable_bias": true,
    "eos_token_id": 2,
    "ffn_dim": 3072,
    "hidden_size": 768,
    "init_std": 0.02,
    "layer_norm_elementwise_affine": true,
    "layerdrop": 0.0,
    "max_position_embeddings": 2048,
    "model_type": "qalb",
    "num_attention_heads": 12,
    "num_hidden_layers": 12,
    "pad_token_id": 1,
    "prefix": "</s>",
    "transformers_version": "4.57.3",
    "use_cache": true,
    "vocab_size": 50272,
    "word_embed_proj_dim": 768,
    "auto_map": {
        "AutoConfig": "modeling_qalb.QalbConfig",
        "AutoModelForCausalLM": "modeling_qalb.FinalPerfectQalb"
    }
}