ChillPower/new_loftr_model
08
1{2 "architectures": [3 "LlamaForCausalLM"4 ],5 "attention_bias": false,6 "attention_dropout": 0.0,7 "bos_token_id": 128000,8 "eos_token_id": 128001,9 "head_dim": 128,10 "hidden_act": "silu",11 "hidden_size": 4096,12 "initializer_range": 0.02,13 "intermediate_size": 14336,14 "max_position_embeddings": 131072,15 "mlp_bias": false,16 "model_type": "llama",17 "num_attention_heads": 32,18 "num_hidden_layers": 32,19 "num_key_value_heads": 8,20 "pad_token_id": 128004,21 "pretraining_tp": 1,22 "quantization_config": {23 "bnb_4bit_compute_dtype": "float16",24 "bnb_4bit_quant_type": "nf4",25 "bnb_4bit_use_double_quant": true,26 "llm_int8_enable_fp32_cpu_offload": false,27 "llm_int8_has_fp16_weight": false,28 "llm_int8_skip_modules": [29 "model.layers.1.mlp.gate_proj",30 "model.layers.1.mlp.up_proj",31 "model.layers.1.mlp.down_proj",32 "lm_head"33 ],34 "llm_int8_threshold": 6.0,35 "load_in_4bit": true,36 "load_in_8bit": false,37 "quant_method": "bitsandbytes"38 },39 "rms_norm_eps": 1e-05,40 "rope_scaling": {41 "factor": 8.0,42 "high_freq_factor": 4.0,43 "low_freq_factor": 1.0,44 "original_max_position_embeddings": 8192,45 "rope_type": "llama3"46 },47 "rope_theta": 500000.0,48 "tie_word_embeddings": false,49 "torch_dtype": "float16",50 "transformers_version": "4.55.2",51 "unsloth_fixed": true,52 "unsloth_version": "2025.8.9",53 "use_cache": true,54 "vocab_size": 12825655}56 