Ring-1T-preview-3bit / config.json
awni's picture
Add files using upload-large-folder tool
89f1f60 verified
{
"architectures": [
"BailingMoeV2ForCausalLM"
],
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "configuration_bailing_moe_v2.BailingMoeV2Config",
"AutoModel": "modeling_bailing_moe_v2.BailingMoeV2Model",
"AutoModelForCausalLM": "modeling_bailing_moe_v2.BailingMoeV2ForCausalLM"
},
"embedding_dropout": 0.0,
"eos_token_id": 156892,
"first_k_dense_replace": 4,
"head_dim": 128,
"hidden_act": "silu",
"hidden_size": 8192,
"intermediate_size": 18432,
"max_position_embeddings": 32768,
"model_type": "bailing_moe",
"moe_intermediate_size": 2048,
"moe_router_enable_expert_bias": true,
"n_group": 8,
"norm_head": false,
"norm_softmax": false,
"norm_topk_prob": true,
"num_attention_heads": 64,
"num_experts": 256,
"num_experts_per_tok": 8,
"num_hidden_layers": 80,
"num_key_value_heads": 8,
"num_shared_experts": 1,
"output_dropout": 0.0,
"pad_token_id": 156892,
"quantization": {
"group_size": 64,
"bits": 3,
"mode": "affine",
"model.layers.4.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.5.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.6.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.7.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.8.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.9.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.10.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.11.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.12.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.13.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.14.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.15.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.16.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.17.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.18.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.19.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.20.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.21.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.22.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.23.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.24.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.25.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.26.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.27.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.28.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.29.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.30.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.31.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.32.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.33.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.34.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.35.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.36.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.37.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.38.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.39.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.40.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.41.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.42.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.43.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.44.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.45.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.46.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.47.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.48.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.49.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.50.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.51.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.52.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.53.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.54.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.55.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.56.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.57.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.58.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.59.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.60.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.61.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.62.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.63.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.64.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.65.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.66.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.67.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.68.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.69.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.70.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.71.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.72.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.73.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.74.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.75.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.76.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.77.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.78.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.79.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
}
},
"quantization_config": {
"group_size": 64,
"bits": 3,
"mode": "affine",
"model.layers.4.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.5.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.6.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.7.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.8.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.9.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.10.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.11.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.12.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.13.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.14.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.15.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.16.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.17.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.18.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.19.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.20.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.21.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.22.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.23.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.24.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.25.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.26.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.27.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.28.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.29.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.30.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.31.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.32.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.33.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.34.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.35.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.36.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.37.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.38.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.39.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.40.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.41.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.42.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.43.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.44.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.45.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.46.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.47.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.48.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.49.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.50.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.51.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.52.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.53.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.54.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.55.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.56.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.57.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.58.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.59.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.60.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.61.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.62.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.63.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.64.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.65.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.66.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.67.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.68.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.69.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.70.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.71.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.72.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.73.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.74.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.75.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.76.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.77.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.78.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
},
"model.layers.79.mlp.gate.gate_proj": {
"group_size": 64,
"bits": 8
}
},
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 600000,
"rotary_dim": 64,
"routed_scaling_factor": 2.5,
"router_dtype": "fp32",
"score_function": "sigmoid",
"tie_word_embeddings": false,
"topk_group": 4,
"torch_dtype": "bfloat16",
"transformers_version": "4.52.3",
"use_bias": false,
"use_cache": true,
"use_qk_norm": true,
"use_qkv_bias": false,
"use_rmsnorm": true,
"using_split_qkv_in_self_attention": false,
"vocab_size": 157184
}