| { |
| "architectures": [ |
| "GlmImageForConditionalGeneration" |
| ], |
| "dtype": "bfloat16", |
| "image_end_token_id": 16385, |
| "image_start_token_id": 16384, |
| "image_token_id": 167855, |
| "model_type": "glm_image", |
| "quantization_config": { |
| "_load_in_4bit": true, |
| "_load_in_8bit": false, |
| "bnb_4bit_compute_dtype": "bfloat16", |
| "bnb_4bit_quant_storage": "uint8", |
| "bnb_4bit_quant_type": "nf4", |
| "bnb_4bit_use_double_quant": true, |
| "llm_int8_enable_fp32_cpu_offload": false, |
| "llm_int8_has_fp16_weight": false, |
| "llm_int8_skip_modules": [ |
| "vqmodel", |
| "lm_head", |
| "embeddings", |
| "embed_tokens", |
| "patch_embed", |
| "model.language_model.layers.28.self_attn.v_proj", |
| "model.language_model.layers.29.self_attn.v_proj", |
| "model.language_model.layers.32.self_attn.v_proj", |
| "model.language_model.layers.33.self_attn.v_proj", |
| "model.visual.blocks.0.attn.qkv", |
| "model.visual.blocks.0.mlp.fc2", |
| "model.visual.blocks.1.attn.qkv", |
| "model.visual.blocks.1.mlp.fc1", |
| "model.visual.blocks.1.mlp.fc2", |
| "model.visual.blocks.2.attn.qkv", |
| "model.visual.blocks.2.mlp.fc2", |
| "model.visual.blocks.3.attn.qkv", |
| "model.visual.blocks.3.mlp.fc2", |
| "model.visual.blocks.4.mlp.fc2", |
| "model.visual.blocks.5.mlp.fc2", |
| "model.visual.blocks.6.mlp.fc2", |
| "model.visual.blocks.7.mlp.fc2", |
| "model.visual.blocks.8.mlp.fc2", |
| "model.visual.blocks.9.mlp.fc2" |
| ], |
| "llm_int8_threshold": 6.0, |
| "load_in_4bit": true, |
| "load_in_8bit": false, |
| "quant_method": "bitsandbytes" |
| }, |
| "text_config": { |
| "attention_bias": true, |
| "attention_dropout": 0.0, |
| "dtype": "bfloat16", |
| "eos_token_id": 16385, |
| "hidden_act": "silu", |
| "hidden_size": 4096, |
| "initializer_range": 0.02, |
| "intermediate_size": 13696, |
| "max_position_embeddings": 131072, |
| "model_type": "glm_image_text", |
| "num_attention_heads": 32, |
| "num_hidden_layers": 40, |
| "num_key_value_heads": 2, |
| "pad_token_id": 167841, |
| "rms_norm_eps": 1e-05, |
| "rope_parameters": { |
| "mrope_section": [ |
| 8, |
| 12, |
| 12 |
| ], |
| "partial_rotary_factor": 0.5, |
| "rope_theta": 10000, |
| "rope_type": "default" |
| }, |
| "use_cache": true, |
| "vision_vocab_size": 16512, |
| "vocab_size": 168064 |
| }, |
| "tie_word_embeddings": false, |
| "transformers_version": "5.0.0", |
| "vision_config": { |
| "attention_bias": true, |
| "attention_dropout": 0.0, |
| "depth": 40, |
| "dtype": "bfloat16", |
| "hidden_act": "gelu", |
| "hidden_size": 1536, |
| "image_size": 2048, |
| "in_channels": 3, |
| "initializer_range": 0.02, |
| "intermediate_size": 6144, |
| "layer_norm_eps": 1e-06, |
| "model_type": "glm_image_vision", |
| "num_heads": 16, |
| "patch_size": 16, |
| "spatial_merge_size": 1 |
| }, |
| "vq_config": { |
| "dtype": "bfloat16", |
| "embed_dim": 2048, |
| "in_channels": 3, |
| "initializer_range": 0.02, |
| "latent_channels": 1536, |
| "model_type": "glm_image_vqmodel", |
| "num_embeddings": 16384 |
| } |
| } |
|
|
