diff --git "a/config.json" "b/config.json" new file mode 100644--- /dev/null +++ "b/config.json" @@ -0,0 +1,5108 @@ +{ + "architectures": [ + "Qwen3_5ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "attn_output_gate": true, + "bos_token_id": 248044, + "dtype": "bfloat16", + "eos_token_id": [ + 248046, + 248044 + ], + "full_attention_interval": 4, + "head_dim": 256, + "hidden_act": "silu", + "hidden_size": 5120, + "initializer_range": 0.02, + "intermediate_size": 17408, + "layer_types": [ + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention", + "linear_attention", + "linear_attention", + "linear_attention", + "full_attention" + ], + "linear_conv_kernel_dim": 4, + "linear_key_head_dim": 128, + "linear_num_key_heads": 16, + "linear_num_value_heads": 48, + "linear_value_head_dim": 128, + "mamba_ssm_dtype": "float32", + "max_position_embeddings": 262144, + "model_type": "qwen3_5", + "mtp_num_hidden_layers": 1, + "mtp_use_dedicated_embeddings": false, + "num_attention_heads": 24, + "num_hidden_layers": 64, + "num_key_value_heads": 4, + "output_gate_type": "swish", + "pad_token_id": null, + "partial_rotary_factor": 0.25, + "quantization": { + "group_size": 64, + "bits": 4, + "mode": "affine", + "language_model.model.embed_tokens": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.0.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.1.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.2.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.3.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.4.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.5.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.6.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.7.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.10.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.13.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.16.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.19.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.22.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.25.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.28.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.31.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.34.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.37.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.40.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.43.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.46.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.49.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.52.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.55.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.56.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.57.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.58.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.59.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.60.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.61.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.62.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.63.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.lm_head": { + "group_size": null, + "bits": 6, + "mode": "affine" + } + }, + "quantization_config": { + "group_size": 64, + "bits": 4, + "mode": "affine", + "language_model.model.embed_tokens": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.0.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.0.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.1.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.1.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.2.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.2.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.3.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.3.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.3.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.4.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.4.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.5.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.5.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.6.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.6.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.7.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.7.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.7.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.8.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.9.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.10.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.10.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.11.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.12.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.13.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.13.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.14.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.15.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.16.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.16.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.17.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.18.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.19.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.19.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.19.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.20.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.21.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.22.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.22.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.23.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.24.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.25.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.25.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.26.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.27.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.28.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.28.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.29.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.30.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.31.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.31.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.31.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.32.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.33.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.34.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.34.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.35.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.36.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.37.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.37.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.38.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.39.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.40.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.40.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.41.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.42.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.43.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.43.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.43.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.44.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.45.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.46.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.46.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.47.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.48.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.49.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.49.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.50.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.v_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.51.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.52.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.52.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.53.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.mlp.down_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.54.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.55.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.55.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.55.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.56.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.56.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.57.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.57.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.58.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.58.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.59.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.59.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.59.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.60.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.60.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.61.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.61.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_qkv": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_z": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_b": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.in_proj_a": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.linear_attn.out_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.62.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.62.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.q_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.k_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.v_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.63.self_attn.o_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.mlp.gate_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.model.layers.63.mlp.down_proj": { + "group_size": null, + "bits": 6, + "mode": "affine" + }, + "language_model.model.layers.63.mlp.up_proj": { + "group_size": null, + "bits": 3, + "mode": "affine" + }, + "language_model.lm_head": { + "group_size": null, + "bits": 6, + "mode": "affine" + } + }, + "rms_norm_eps": 1e-06, + "rope_parameters": { + "mrope_interleaved": true, + "mrope_section": [ + 11, + 11, + 10 + ], + "partial_rotary_factor": 0.25, + "rope_theta": 10000000, + "type": "default" + }, + "tie_word_embeddings": false, + "transformers_version": "5.7.0.dev0", + "use_cache": true, + "vocab_size": 248320 +} \ No newline at end of file