QuantTrio/KAT-Dev-GPTQ-Int4

1{2  "architectures": [3    "Qwen3ForCausalLM"4  ],5  "attention_bias": false,6  "attention_dropout": 0.0,7  "dtype": "bfloat16",8  "eos_token_id": 151645,9  "head_dim": 128,10  "hidden_act": "silu",11  "hidden_size": 5120,12  "initializer_range": 0.02,13  "intermediate_size": 25600,14  "layer_types": [15    "full_attention",16    "full_attention",17    "full_attention",18    "full_attention",19    "full_attention",20    "full_attention",21    "full_attention",22    "full_attention",23    "full_attention",24    "full_attention",25    "full_attention",26    "full_attention",27    "full_attention",28    "full_attention",29    "full_attention",30    "full_attention",31    "full_attention",32    "full_attention",33    "full_attention",34    "full_attention",35    "full_attention",36    "full_attention",37    "full_attention",38    "full_attention",39    "full_attention",40    "full_attention",41    "full_attention",42    "full_attention",43    "full_attention",44    "full_attention",45    "full_attention",46    "full_attention",47    "full_attention",48    "full_attention",49    "full_attention",50    "full_attention",51    "full_attention",52    "full_attention",53    "full_attention",54    "full_attention",55    "full_attention",56    "full_attention",57    "full_attention",58    "full_attention",59    "full_attention",60    "full_attention",61    "full_attention",62    "full_attention",63    "full_attention",64    "full_attention",65    "full_attention",66    "full_attention",67    "full_attention",68    "full_attention",69    "full_attention",70    "full_attention",71    "full_attention",72    "full_attention",73    "full_attention",74    "full_attention",75    "full_attention",76    "full_attention",77    "full_attention",78    "full_attention"79  ],80  "max_position_embeddings": 131072,81  "max_window_layers": 64,82  "model_type": "qwen3",83  "num_attention_heads": 64,84  "num_hidden_layers": 64,85  "num_key_value_heads": 8,86  "pad_token_id": 151643,87  "quantization_config": {88    "bits": 4,89    "checkpoint_format": "gptq",90    "desc_act": false,91    "group_size": 128,92    "lm_head": false,93    "meta": {94      "act_group_aware": false,95      "damp_auto_increment": 0.01,96      "damp_percent": 0.05,97      "mse": 0.0,98      "quantizer": [99        "gptqmodel:4.2.5"100      ],101      "static_groups": false,102      "true_sequential": true,103      "uri": "https://github.com/modelcloud/gptqmodel",104      "v2": false,105      "v2_alpha": 0.25106    },107    "pack_dtype": "int32",108    "quant_method": "gptq",109    "sym": true110  },111  "rms_norm_eps": 1e-06,112  "rope_scaling": null,113  "rope_theta": 1000000,114  "sliding_window": null,115  "tie_word_embeddings": false,116  "transformers_version": "4.56.2",117  "use_cache": false,118  "use_sliding_window": false,119  "vocab_size": 151936120}121