FenomAI/North-Mini-Code-1.0-AWQ-INT4

1{2  "architectures": [3    "Cohere2MoeForCausalLM"4  ],5  "attention_bias": false,6  "attention_dropout": 0.0,7  "bos_token_id": 2,8  "dtype": "float16",9  "eos_token_id": 255001,10  "expert_selection_fn": "sigmoid",11  "first_k_dense_replace": 1,12  "head_dim": 128,13  "hidden_act": "silu",14  "hidden_size": 2048,15  "initializer_range": 0.02,16  "intermediate_size": 768,17  "layer_norm_eps": 1e-05,18  "layer_types": [19    "full_attention",20    "sliding_attention",21    "sliding_attention",22    "sliding_attention",23    "full_attention",24    "sliding_attention",25    "sliding_attention",26    "sliding_attention",27    "full_attention",28    "sliding_attention",29    "sliding_attention",30    "sliding_attention",31    "full_attention",32    "sliding_attention",33    "sliding_attention",34    "sliding_attention",35    "full_attention",36    "sliding_attention",37    "sliding_attention",38    "sliding_attention",39    "full_attention",40    "sliding_attention",41    "sliding_attention",42    "sliding_attention",43    "full_attention",44    "sliding_attention",45    "sliding_attention",46    "sliding_attention",47    "full_attention",48    "sliding_attention",49    "sliding_attention",50    "sliding_attention",51    "full_attention",52    "sliding_attention",53    "sliding_attention",54    "sliding_attention",55    "full_attention",56    "sliding_attention",57    "sliding_attention",58    "sliding_attention",59    "full_attention",60    "sliding_attention",61    "sliding_attention",62    "sliding_attention",63    "full_attention",64    "sliding_attention",65    "sliding_attention",66    "sliding_attention",67    "full_attention"68  ],69  "logit_scale": 1.0,70  "max_position_embeddings": 500000,71  "mlp_layer_types": [72    "dense",73    "sparse",74    "sparse",75    "sparse",76    "sparse",77    "sparse",78    "sparse",79    "sparse",80    "sparse",81    "sparse",82    "sparse",83    "sparse",84    "sparse",85    "sparse",86    "sparse",87    "sparse",88    "sparse",89    "sparse",90    "sparse",91    "sparse",92    "sparse",93    "sparse",94    "sparse",95    "sparse",96    "sparse",97    "sparse",98    "sparse",99    "sparse",100    "sparse",101    "sparse",102    "sparse",103    "sparse",104    "sparse",105    "sparse",106    "sparse",107    "sparse",108    "sparse",109    "sparse",110    "sparse",111    "sparse",112    "sparse",113    "sparse",114    "sparse",115    "sparse",116    "sparse",117    "sparse",118    "sparse",119    "sparse",120    "sparse"121  ],122  "model_type": "cohere2_moe",123  "norm_topk_prob": false,124  "num_attention_heads": 32,125  "num_experts": 128,126  "num_experts_per_tok": 8,127  "num_hidden_layers": 49,128  "num_key_value_heads": 4,129  "num_shared_experts": 0,130  "pad_token_id": 0,131  "prefix_dense_intermediate_size": 3072,132  "prefix_dense_sliding_window_pattern": 1,133  "quantization_config": {134    "config_groups": {135      "group_0": {136        "format": "pack-quantized",137        "input_activations": null,138        "output_activations": null,139        "targets": [140          "Linear"141        ],142        "weights": {143          "actorder": null,144          "block_structure": null,145          "dynamic": false,146          "group_size": 32,147          "num_bits": 4,148          "observer": "mse",149          "observer_kwargs": {},150          "scale_dtype": null,151          "strategy": "group",152          "symmetric": false,153          "type": "int",154          "zp_dtype": "torch.int8"155        }156      }157    },158    "format": "pack-quantized",159    "global_compression_ratio": null,160    "ignore": [161      "model.layers.0.self_attn.q_proj",162      "model.layers.0.self_attn.k_proj",163      "model.layers.0.self_attn.v_proj",164      "model.layers.0.self_attn.o_proj",165      "model.layers.0.mlp.gate_proj",166      "model.layers.0.mlp.up_proj",167      "model.layers.0.mlp.down_proj",168      "lm_head",169      "re:.*mlp[.]gate"170    ],171    "kv_cache_scheme": null,172    "quant_method": "compressed-tensors",173    "quantization_status": "compressed",174    "sparsity_config": {},175    "transform_config": {},176    "version": "0.1.dev489+g6d69620"177  },178  "rms_norm_eps": 1e-06,179  "rope_parameters": {180    "rope_theta": 50000,181    "rope_type": "default"182  },183  "rope_scaling": null,184  "rope_theta": 50000,185  "shared_expert_combination_strategy": "average",186  "sliding_window": 4096,187  "sliding_window_pattern": 4,188  "tie_word_embeddings": true,189  "transformers_version": "5.10.2",190  "use_cache": true,191  "use_gated_activation": true,192  "use_parallel_block": true,193  "use_parallel_embedding": false,194  "use_qk_norm": false,195  "vocab_size": 262144196}