jhonacmarvik/distilbert-command-classifier_quantized

1{2  "activation": "gelu",3  "architectures": [4    "DistilBertForSequenceClassification"5  ],6  "attention_dropout": 0.1,7  "dim": 768,8  "dropout": 0.1,9  "dtype": "float16",10  "hidden_dim": 3072,11  "id2label": {12    "0": "command",13    "1": "question"14  },15  "initializer_range": 0.02,16  "label2id": {17    "command": 0,18    "question": 119  },20  "max_position_embeddings": 512,21  "model_type": "distilbert",22  "n_heads": 12,23  "n_layers": 6,24  "pad_token_id": 0,25  "problem_type": "single_label_classification",26  "qa_dropout": 0.1,27  "quantization_config": {28    "_load_in_4bit": false,29    "_load_in_8bit": true,30    "bnb_4bit_compute_dtype": "float32",31    "bnb_4bit_quant_storage": "uint8",32    "bnb_4bit_quant_type": "fp4",33    "bnb_4bit_use_double_quant": false,34    "llm_int8_enable_fp32_cpu_offload": false,35    "llm_int8_has_fp16_weight": false,36    "llm_int8_skip_modules": null,37    "llm_int8_threshold": 6.0,38    "load_in_4bit": false,39    "load_in_8bit": true,40    "quant_method": "bitsandbytes"41  },42  "seq_classif_dropout": 0.2,43  "sinusoidal_pos_embds": false,44  "tie_weights_": true,45  "transformers_version": "4.57.3",46  "vocab_size": 3052247}48