AnwarWall-e/profiler
011
1{2 "architectures": [3 "LlamaForCausalLM"4 ],5 "attention_bias": false,6 "attention_dropout": 0.0,7 "bos_token_id": 128000,8 "dtype": "float16",9 "eos_token_id": 128009,10 "head_dim": 64,11 "hidden_act": "silu",12 "hidden_size": 2048,13 "initializer_range": 0.02,14 "intermediate_size": 8192,15 "max_position_embeddings": 131072,16 "mlp_bias": false,17 "model_type": "llama",18 "num_attention_heads": 32,19 "num_hidden_layers": 16,20 "num_key_value_heads": 8,21 "pad_token_id": 128004,22 "pretraining_tp": 1,23 "quantization_config": {24 "bnb_4bit_compute_dtype": "float16",25 "bnb_4bit_quant_type": "nf4",26 "bnb_4bit_use_double_quant": true,27 "llm_int8_enable_fp32_cpu_offload": false,28 "llm_int8_has_fp16_weight": false,29 "llm_int8_skip_modules": null,30 "llm_int8_threshold": 6.0,31 "load_in_4bit": true,32 "load_in_8bit": false,33 "quant_method": "bitsandbytes"34 },35 "rms_norm_eps": 1e-05,36 "rope_scaling": {37 "factor": 32.0,38 "high_freq_factor": 4.0,39 "low_freq_factor": 1.0,40 "original_max_position_embeddings": 8192,41 "rope_type": "llama3"42 },43 "rope_theta": 500000.0,44 "tie_word_embeddings": true,45 "transformers_version": "4.57.3",46 "unsloth_fixed": true,47 "unsloth_version": "2025.12.5",48 "use_cache": true,49 "vocab_size": 12825650}51 