virtualvoidsteve/code_correction_classifier_model_2305
010
1{2 "_name_or_path": "google/gemma-7b",3 "architectures": [4 "GemmaForSequenceClassification"5 ],6 "attention_bias": false,7 "attention_dropout": 0.0,8 "bos_token_id": 2,9 "eos_token_id": 1,10 "head_dim": 256,11 "hidden_act": "gelu",12 "hidden_activation": null,13 "hidden_size": 3072,14 "initializer_range": 0.02,15 "intermediate_size": 24576,16 "max_position_embeddings": 8192,17 "model_type": "gemma",18 "num_attention_heads": 16,19 "num_hidden_layers": 28,20 "num_key_value_heads": 16,21 "pad_token_id": 0,22 "pretraining_tp": 1,23 "quantization_config": {24 "_load_in_4bit": true,25 "_load_in_8bit": false,26 "bnb_4bit_compute_dtype": "float16",27 "bnb_4bit_quant_storage": "uint8",28 "bnb_4bit_quant_type": "nf4",29 "bnb_4bit_use_double_quant": false,30 "llm_int8_enable_fp32_cpu_offload": false,31 "llm_int8_has_fp16_weight": false,32 "llm_int8_skip_modules": null,33 "llm_int8_threshold": 6.0,34 "load_in_4bit": true,35 "load_in_8bit": false,36 "quant_method": "bitsandbytes"37 },38 "rms_norm_eps": 1e-06,39 "rope_scaling": null,40 "rope_theta": 10000.0,41 "torch_dtype": "float32",42 "transformers_version": "4.40.0.dev0",43 "use_cache": true,44 "vocab_size": 25600045}46 