diff --git "a/logs.txt" "b/logs.txt" new file mode 100644--- /dev/null +++ "b/logs.txt" @@ -0,0 +1,751 @@ +/opt/conda/envs/py310/bin/python -m mlc_llm gen_config /models/Qwen2-0.5B-Instruct --quantization q4f32_1 --conv-template chatml --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +[2024-06-06 23:38:15] INFO auto_config.py:116: Found model configuration: /models/Qwen2-0.5B-Instruct/config.json +[2024-06-06 23:38:15] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override. +[2024-06-06 23:38:15] INFO qwen2_model.py:49: context_window_size not found in config.json. Falling back to max_position_embeddings (32768) +[2024-06-06 23:38:15] INFO qwen2_model.py:66: prefill_chunk_size defaults to 2048 +[2024-06-06 23:38:15] INFO config.py:107: Overriding max_batch_size from 1 to 80 +[2024-06-06 23:38:15] INFO gen_config.py:143: [generation_config.json] Setting bos_token_id: 151643 +[2024-06-06 23:38:15] INFO gen_config.py:143: [generation_config.json] Setting pad_token_id: 151643 +[2024-06-06 23:38:15] INFO gen_config.py:143: [generation_config.json] Setting eos_token_id: [151645, 151643] +[2024-06-06 23:38:15] INFO gen_config.py:143: [generation_config.json] Setting repetition_penalty: 1.1 +[2024-06-06 23:38:15] INFO gen_config.py:143: [generation_config.json] Setting temperature: 0.7 +[2024-06-06 23:38:15] INFO gen_config.py:143: [generation_config.json] Setting top_p: 0.8 +[2024-06-06 23:38:15] INFO gen_config.py:157: Not found tokenizer config: /models/Qwen2-0.5B-Instruct/tokenizer.model +[2024-06-06 23:38:15] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-0.5B-Instruct/tokenizer.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/tokenizer.json +[2024-06-06 23:38:15] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-0.5B-Instruct/vocab.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/vocab.json +[2024-06-06 23:38:15] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-0.5B-Instruct/merges.txt. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/merges.txt +[2024-06-06 23:38:15] INFO gen_config.py:157: Not found tokenizer config: /models/Qwen2-0.5B-Instruct/added_tokens.json +[2024-06-06 23:38:15] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-0.5B-Instruct/tokenizer_config.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/tokenizer_config.json +[2024-06-06 23:38:15] INFO gen_config.py:216: Detected tokenizer info: {'token_postproc_method': 'byte_level', 'prepend_space_in_encode': False, 'strip_space_in_decode': False} +[2024-06-06 23:38:15] INFO gen_config.py:32: [System default] Setting presence_penalty: 0.0 +[2024-06-06 23:38:15] INFO gen_config.py:32: [System default] Setting frequency_penalty: 0.0 +[2024-06-06 23:38:15] INFO gen_config.py:223: Dumping configuration file to: /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/mlc-chat-config.json +/opt/conda/envs/py310/bin/python -m mlc_llm convert_weight /models/Qwen2-0.5B-Instruct --quantization q4f32_1 --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +[2024-06-06 23:38:17] INFO auto_config.py:116: Found model configuration: /models/Qwen2-0.5B-Instruct/config.json +[2024-06-06 23:38:18] INFO auto_device.py:79: Found device: cuda:0 +[2024-06-06 23:38:20] INFO auto_device.py:88: Not found device: rocm:0 +[2024-06-06 23:38:21] INFO auto_device.py:88: Not found device: metal:0 +[2024-06-06 23:38:23] INFO auto_device.py:79: Found device: vulkan:0 +[2024-06-06 23:38:23] INFO auto_device.py:79: Found device: vulkan:1 +[2024-06-06 23:38:23] INFO auto_device.py:79: Found device: vulkan:2 +[2024-06-06 23:38:23] INFO auto_device.py:79: Found device: vulkan:3 +[2024-06-06 23:38:24] INFO auto_device.py:88: Not found device: opencl:0 +[2024-06-06 23:38:24] INFO auto_device.py:35: Using device: cuda:0 +[2024-06-06 23:38:24] INFO auto_weight.py:71: Finding weights in: /models/Qwen2-0.5B-Instruct +[2024-06-06 23:38:24] INFO auto_weight.py:137: Not found Huggingface PyTorch +[2024-06-06 23:38:24] INFO auto_weight.py:144: Found source weight format: huggingface-safetensor. Source configuration: /models/Qwen2-0.5B-Instruct/model.safetensors.index.json +[2024-06-06 23:38:24] INFO auto_weight.py:107: Using source weight configuration: /models/Qwen2-0.5B-Instruct/model.safetensors.index.json. Use `--source` to override. +[2024-06-06 23:38:24] INFO auto_weight.py:111: Using source weight format: huggingface-safetensor. Use `--source-format` to override. +[2024-06-06 23:38:24] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override. +[2024-06-06 23:38:24] INFO qwen2_model.py:49: context_window_size not found in config.json. Falling back to max_position_embeddings (32768) +[2024-06-06 23:38:24] INFO qwen2_model.py:66: prefill_chunk_size defaults to 2048 +Traceback (most recent call last): + File "/opt/conda/envs/py310/lib/python3.10/runpy.py", line 196, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/envs/py310/lib/python3.10/runpy.py", line 86, in _run_code + exec(code, run_globals) + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/__main__.py", line 64, in + main() + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/__main__.py", line 37, in main + cli.main(sys.argv[2:]) + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/cli/convert_weight.py", line 88, in main + convert_weight( + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/interface/convert_weight.py", line 181, in convert_weight + _convert_args(args) + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/interface/convert_weight.py", line 145, in _convert_args + tvmjs.dump_ndarray_cache( + File "/opt/conda/envs/py310/lib/python3.10/site-packages/tvm/contrib/tvmjs.py", line 272, in dump_ndarray_cache + for k, origin_v in param_generator: + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/interface/convert_weight.py", line 122, in _param_generator + loader = LOADER[args.source_format]( + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/loader/huggingface_loader.py", line 99, in __init__ + check_parameter_usage(extern_param_map, set(self.torch_to_path.keys())) + File "/opt/conda/envs/py310/lib/python3.10/site-packages/mlc_llm/loader/utils.py", line 33, in check_parameter_usage + raise ValueError( +ValueError: The following extern parameters do not exist in the weight files: + lm_head.weight +Weight conversion with arguments: + --config /models/Qwen2-0.5B-Instruct/config.json + --quantization GroupQuantize(name='q4f32_1', kind='group-quant', group_size=32, quantize_dtype='int4', storage_dtype='uint32', model_dtype='float32', linear_weight_layout='NK', quantize_embedding=True, quantize_final_fc=True, num_elem_per_storage=8, num_storage_per_group=4, max_int_value=7) + --model-type qwen2 + --device cuda:0 + --source /models/Qwen2-0.5B-Instruct/model.safetensors.index.json + --source-format huggingface-safetensor + --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +Start storing to cache /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +/home/rickzhou/miniconda3/envs/mlc/bin/python -m mlc_llm gen_config /ssd2/models/Qwen2-0.5B-Instruct --quantization q4f32_1 --conv-template chatml --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +[2024-06-06 22:34:16] INFO auto_config.py:116: Found model configuration: /ssd2/models/Qwen2-0.5B-Instruct/config.json +[2024-06-06 22:34:16] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override. +[2024-06-06 22:34:16] INFO qwen2_model.py:50: context_window_size not found in config.json. Falling back to max_position_embeddings (32768) +[2024-06-06 22:34:16] INFO qwen2_model.py:67: prefill_chunk_size defaults to 2048 +[2024-06-06 22:34:16] INFO config.py:107: Overriding max_batch_size from 1 to 80 +[2024-06-06 22:34:16] INFO gen_config.py:143: [generation_config.json] Setting bos_token_id: 151643 +[2024-06-06 22:34:16] INFO gen_config.py:143: [generation_config.json] Setting pad_token_id: 151643 +[2024-06-06 22:34:16] INFO gen_config.py:143: [generation_config.json] Setting eos_token_id: [151645, 151643] +[2024-06-06 22:34:16] INFO gen_config.py:143: [generation_config.json] Setting repetition_penalty: 1.1 +[2024-06-06 22:34:16] INFO gen_config.py:143: [generation_config.json] Setting temperature: 0.7 +[2024-06-06 22:34:16] INFO gen_config.py:143: [generation_config.json] Setting top_p: 0.8 +[2024-06-06 22:34:16] INFO gen_config.py:157: Not found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/tokenizer.model +[2024-06-06 22:34:16] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/tokenizer.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/tokenizer.json +[2024-06-06 22:34:16] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/vocab.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/vocab.json +[2024-06-06 22:34:16] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/merges.txt. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/merges.txt +[2024-06-06 22:34:16] INFO gen_config.py:157: Not found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/added_tokens.json +[2024-06-06 22:34:16] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/tokenizer_config.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/tokenizer_config.json +[2024-06-06 22:34:16] INFO gen_config.py:216: Detected tokenizer info: {'token_postproc_method': 'byte_level', 'prepend_space_in_encode': False, 'strip_space_in_decode': False} +[2024-06-06 22:34:16] INFO gen_config.py:32: [System default] Setting presence_penalty: 0.0 +[2024-06-06 22:34:16] INFO gen_config.py:32: [System default] Setting frequency_penalty: 0.0 +[2024-06-06 22:34:16] INFO gen_config.py:223: Dumping configuration file to: /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/mlc-chat-config.json +/home/rickzhou/miniconda3/envs/mlc/bin/python -m mlc_llm convert_weight /ssd2/models/Qwen2-0.5B-Instruct --quantization q4f32_1 --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +[2024-06-06 22:34:17] INFO auto_config.py:116: Found model configuration: /ssd2/models/Qwen2-0.5B-Instruct/config.json +[2024-06-06 22:34:18] INFO auto_device.py:79: Found device: cuda:0 +[2024-06-06 22:34:18] INFO auto_device.py:79: Found device: cuda:1 +[2024-06-06 22:34:19] INFO auto_device.py:88: Not found device: rocm:0 +[2024-06-06 22:34:21] INFO auto_device.py:88: Not found device: metal:0 +[2024-06-06 22:34:22] INFO auto_device.py:79: Found device: vulkan:0 +[2024-06-06 22:34:22] INFO auto_device.py:79: Found device: vulkan:1 +[2024-06-06 22:34:22] INFO auto_device.py:79: Found device: vulkan:2 +[2024-06-06 22:34:23] INFO auto_device.py:88: Not found device: opencl:0 +[2024-06-06 22:34:23] INFO auto_device.py:35: Using device: cuda:0 +[2024-06-06 22:34:23] INFO auto_weight.py:71: Finding weights in: /ssd2/models/Qwen2-0.5B-Instruct +[2024-06-06 22:34:23] INFO auto_weight.py:137: Not found Huggingface PyTorch +[2024-06-06 22:34:23] INFO auto_weight.py:144: Found source weight format: huggingface-safetensor. Source configuration: /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors.index.json +[2024-06-06 22:34:23] INFO auto_weight.py:107: Using source weight configuration: /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors.index.json. Use `--source` to override. +[2024-06-06 22:34:23] INFO auto_weight.py:111: Using source weight format: huggingface-safetensor. Use `--source-format` to override. +[2024-06-06 22:34:23] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override. +[2024-06-06 22:34:23] INFO qwen2_model.py:50: context_window_size not found in config.json. Falling back to max_position_embeddings (32768) +[2024-06-06 22:34:23] INFO qwen2_model.py:67: prefill_chunk_size defaults to 2048 +Weight conversion with arguments: + --config /ssd2/models/Qwen2-0.5B-Instruct/config.json + --quantization GroupQuantize(name='q4f32_1', kind='group-quant', group_size=32, quantize_dtype='int4', storage_dtype='uint32', model_dtype='float32', linear_weight_layout='NK', quantize_embedding=True, quantize_final_fc=True, num_elem_per_storage=8, num_storage_per_group=4, max_int_value=7) + --model-type qwen2 + --device cuda:0 + --source /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors.index.json + --source-format huggingface-safetensor + --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +Start storing to cache /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC + 0%| | 0/170 [00:00 type is zero. + setattr(self, word, getattr(machar, word).flat[0]) +/home/rickzhou/miniconda3/envs/mlc/lib/python3.11/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. + return self._float_to_str(self.smallest_subnormal) + 1%| | 1/170 [00:03<08:28, 3.01s/it] [2024-06-06 22:34:29] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.input_layernorm.weight", shape: (896,), dtype: float32 + 1%| | 1/170 [00:03<08:28, 3.01s/it] [2024-06-06 22:34:29] INFO group_quantization.py:217: Compiling quantize function for key: ((896, 4864), float32, cuda, axis=1, output_transpose=False) + 1%| | 1/170 [00:03<08:28, 3.01s/it] [2024-06-06 22:34:29] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 1%| | 1/170 [00:03<08:28, 3.01s/it] [2024-06-06 22:34:29] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 1%| | 1/170 [00:03<08:28, 3.01s/it] 2%|▏ | 3/170 [00:03<02:29, 1.11it/s] [2024-06-06 22:34:29] INFO group_quantization.py:217: Compiling quantize function for key: ((9728, 896), float32, cuda, axis=1, output_transpose=False) + 2%|▏ | 3/170 [00:03<02:29, 1.11it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 2%|▏ | 3/170 [00:03<02:29, 1.11it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 2%|▏ | 3/170 [00:03<02:29, 1.11it/s] 2%|▏ | 4/170 [00:03<01:56, 1.43it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 2%|▏ | 4/170 [00:03<01:56, 1.43it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 2%|▏ | 4/170 [00:03<01:56, 1.43it/s] [2024-06-06 22:34:30] INFO group_quantization.py:217: Compiling quantize function for key: ((1152, 896), float32, cuda, axis=1, output_transpose=False) + 2%|▏ | 4/170 [00:03<01:56, 1.43it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 2%|▏ | 4/170 [00:03<01:56, 1.43it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 2%|▏ | 4/170 [00:03<01:56, 1.43it/s] 4%|▍ | 7/170 [00:03<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO group_quantization.py:217: Compiling quantize function for key: ((896, 896), float32, cuda, axis=1, output_transpose=False) + 4%|▍ | 7/170 [00:03<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 4%|▍ | 7/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 4%|▍ | 7/170 [00:04<00:56, 2.88it/s] 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.input_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.input_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:56, 2.88it/s] 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:10, 14.07it/s] 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.input_layernorm.weight", shape: (896,), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.input_layernorm.weight", shape: (896,), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 27%|██▋ | 46/170 [00:04<00:03, 37.59it/s] 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.input_layernorm.weight", shape: (896,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.input_layernorm.weight", shape: (896,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 35%|███▍ | 59/170 [00:04<00:02, 49.51it/s] 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:04<00:01, 66.77it/s] 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:04<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:00, 89.46it/s] 76%|██████���▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 112.96it/s] 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.input_layernorm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] [2024-06-06 22:34:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.norm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 130.72it/s] 100%|██████████| 170/170 [00:05<00:00, 32.09it/s] +[2024-06-06 22:34:31] INFO huggingface_loader.py:197: Unloading HF weight file: /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors +[2024-06-06 22:34:31] INFO stats.py:77: Time usage: HF loading: 1.504 sec; Pre-quantization mapping: 0.357 sec; Quantization: 2.508 sec +[2024-06-06 22:34:31] INFO stats.py:91: RAM usage: Peak RAM: 1.840 GB. Total bytes loaded from disk: 1.840 GB +[2024-06-06 22:34:31] INFO convert_weight.py:155: Parameter size after quantization: 0.288 GB +[2024-06-06 22:34:31] INFO convert_weight.py:160: Total parameters: 494,032,768 +[2024-06-06 22:34:31] INFO convert_weight.py:161: Bits per parameter: 5.004 +[2024-06-06 22:34:31] INFO convert_weight.py:166: Saved to directory: /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC + +All finished, 8 total shards committed, record saved to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/ndarray-cache.json +Also saved a bf16 record to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/ndarray-cache-b16.json +/home/rickzhou/miniconda3/envs/mlc/bin/python -m mlc_llm gen_config /ssd2/models/Qwen2-0.5B-Instruct --quantization q4f32_1 --conv-template chatml --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +[2024-06-06 22:36:45] INFO auto_config.py:116: Found model configuration: /ssd2/models/Qwen2-0.5B-Instruct/config.json +[2024-06-06 22:36:45] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override. +[2024-06-06 22:36:45] INFO qwen2_model.py:50: context_window_size not found in config.json. Falling back to max_position_embeddings (32768) +[2024-06-06 22:36:45] INFO qwen2_model.py:67: prefill_chunk_size defaults to 2048 +[2024-06-06 22:36:45] INFO config.py:107: Overriding max_batch_size from 1 to 80 +[2024-06-06 22:36:45] INFO gen_config.py:143: [generation_config.json] Setting bos_token_id: 151643 +[2024-06-06 22:36:45] INFO gen_config.py:143: [generation_config.json] Setting pad_token_id: 151643 +[2024-06-06 22:36:45] INFO gen_config.py:143: [generation_config.json] Setting eos_token_id: [151645, 151643] +[2024-06-06 22:36:45] INFO gen_config.py:143: [generation_config.json] Setting repetition_penalty: 1.1 +[2024-06-06 22:36:45] INFO gen_config.py:143: [generation_config.json] Setting temperature: 0.7 +[2024-06-06 22:36:45] INFO gen_config.py:143: [generation_config.json] Setting top_p: 0.8 +[2024-06-06 22:36:45] INFO gen_config.py:157: Not found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/tokenizer.model +[2024-06-06 22:36:45] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/tokenizer.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/tokenizer.json +[2024-06-06 22:36:45] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/vocab.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/vocab.json +[2024-06-06 22:36:45] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/merges.txt. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/merges.txt +[2024-06-06 22:36:45] INFO gen_config.py:157: Not found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/added_tokens.json +[2024-06-06 22:36:45] INFO gen_config.py:155: Found tokenizer config: /ssd2/models/Qwen2-0.5B-Instruct/tokenizer_config.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/tokenizer_config.json +[2024-06-06 22:36:45] INFO gen_config.py:216: Detected tokenizer info: {'token_postproc_method': 'byte_level', 'prepend_space_in_encode': False, 'strip_space_in_decode': False} +[2024-06-06 22:36:45] INFO gen_config.py:32: [System default] Setting presence_penalty: 0.0 +[2024-06-06 22:36:45] INFO gen_config.py:32: [System default] Setting frequency_penalty: 0.0 +[2024-06-06 22:36:45] INFO gen_config.py:223: Dumping configuration file to: /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/mlc-chat-config.json +/home/rickzhou/miniconda3/envs/mlc/bin/python -m mlc_llm convert_weight /ssd2/models/Qwen2-0.5B-Instruct --quantization q4f32_1 --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +[2024-06-06 22:36:47] INFO auto_config.py:116: Found model configuration: /ssd2/models/Qwen2-0.5B-Instruct/config.json +[2024-06-06 22:36:48] INFO auto_device.py:79: Found device: cuda:0 +[2024-06-06 22:36:48] INFO auto_device.py:79: Found device: cuda:1 +[2024-06-06 22:36:49] INFO auto_device.py:88: Not found device: rocm:0 +[2024-06-06 22:36:50] INFO auto_device.py:88: Not found device: metal:0 +[2024-06-06 22:36:51] INFO auto_device.py:79: Found device: vulkan:0 +[2024-06-06 22:36:51] INFO auto_device.py:79: Found device: vulkan:1 +[2024-06-06 22:36:51] INFO auto_device.py:79: Found device: vulkan:2 +[2024-06-06 22:36:52] INFO auto_device.py:88: Not found device: opencl:0 +[2024-06-06 22:36:52] INFO auto_device.py:35: Using device: cuda:0 +[2024-06-06 22:36:52] INFO auto_weight.py:71: Finding weights in: /ssd2/models/Qwen2-0.5B-Instruct +[2024-06-06 22:36:52] INFO auto_weight.py:137: Not found Huggingface PyTorch +[2024-06-06 22:36:52] INFO auto_weight.py:144: Found source weight format: huggingface-safetensor. Source configuration: /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors.index.json +[2024-06-06 22:36:52] INFO auto_weight.py:107: Using source weight configuration: /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors.index.json. Use `--source` to override. +[2024-06-06 22:36:52] INFO auto_weight.py:111: Using source weight format: huggingface-safetensor. Use `--source-format` to override. +[2024-06-06 22:36:52] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override. +[2024-06-06 22:36:52] INFO qwen2_model.py:50: context_window_size not found in config.json. Falling back to max_position_embeddings (32768) +[2024-06-06 22:36:52] INFO qwen2_model.py:67: prefill_chunk_size defaults to 2048 +Weight conversion with arguments: + --config /ssd2/models/Qwen2-0.5B-Instruct/config.json + --quantization GroupQuantize(name='q4f32_1', kind='group-quant', group_size=32, quantize_dtype='int4', storage_dtype='uint32', model_dtype='float32', linear_weight_layout='NK', quantize_embedding=True, quantize_final_fc=True, num_elem_per_storage=8, num_storage_per_group=4, max_int_value=7) + --model-type qwen2 + --device cuda:0 + --source /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors.index.json + --source-format huggingface-safetensor + --output /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC +Start storing to cache /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC + 0%| | 0/170 [00:00 type is zero. + setattr(self, word, getattr(machar, word).flat[0]) +/home/rickzhou/miniconda3/envs/mlc/lib/python3.11/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. + return self._float_to_str(self.smallest_subnormal) + 1%| | 1/170 [00:03<08:50, 3.14s/it] [2024-06-06 22:36:58] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.input_layernorm.weight", shape: (896,), dtype: float32 + 1%| | 1/170 [00:03<08:50, 3.14s/it] [2024-06-06 22:36:58] INFO group_quantization.py:217: Compiling quantize function for key: ((896, 4864), float32, cuda, axis=1, output_transpose=False) + 1%| | 1/170 [00:03<08:50, 3.14s/it] [2024-06-06 22:36:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 1%| | 1/170 [00:03<08:50, 3.14s/it] [2024-06-06 22:36:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 1%| | 1/170 [00:03<08:50, 3.14s/it] 2%|▏ | 3/170 [00:03<02:36, 1.07it/s] [2024-06-06 22:36:58] INFO group_quantization.py:217: Compiling quantize function for key: ((9728, 896), float32, cuda, axis=1, output_transpose=False) + 2%|▏ | 3/170 [00:03<02:36, 1.07it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 2%|▏ | 3/170 [00:03<02:36, 1.07it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 2%|▏ | 3/170 [00:03<02:36, 1.07it/s] 2%|▏ | 4/170 [00:03<01:59, 1.38it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 2%|▏ | 4/170 [00:03<01:59, 1.38it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 2%|▏ | 4/170 [00:03<01:59, 1.38it/s] [2024-06-06 22:36:59] INFO group_quantization.py:217: Compiling quantize function for key: ((1152, 896), float32, cuda, axis=1, output_transpose=False) + 2%|▏ | 4/170 [00:03<01:59, 1.38it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 2%|▏ | 4/170 [00:04<01:59, 1.38it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 2%|▏ | 4/170 [00:04<01:59, 1.38it/s] 4%|▍ | 7/170 [00:04<00:57, 2.84it/s] [2024-06-06 22:36:59] INFO group_quantization.py:217: Compiling quantize function for key: ((896, 896), float32, cuda, axis=1, output_transpose=False) + 4%|▍ | 7/170 [00:04<00:57, 2.84it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 4%|▍ | 7/170 [00:04<00:57, 2.84it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 4%|▍ | 7/170 [00:04<00:57, 2.84it/s] 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.input_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.input_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 5%|▍ | 8/170 [00:04<00:55, 2.90it/s] 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.input_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] [2024-06-06 22:36:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 13%|█▎ | 22/170 [00:04<00:11, 13.44it/s] 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.input_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.input_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.input_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.input_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 29%|██▉ | 49/170 [00:04<00:03, 34.69it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 29%|██▉ | 49/170 [00:05<00:03, 34.69it/s] 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.input_layernorm.weight", shape: (896,), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 45%|████▌ | 77/170 [00:05<00:01, 53.87it/s] 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.input_layernorm.weight", shape: (896,), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 60%|██████ | 102/170 [00:05<00:01, 67.30it/s] 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.input_layernorm.weight", shape: (896,), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] [2024-06-06 22:37:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 76%|███████▋ | 130/170 [00:05<00:00, 82.26it/s] 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.input_layernorm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_weight", shape: (896, 608), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_scale", shape: (896, 152), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_weight", shape: (9728, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_scale", shape: (9728, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.post_attention_layernorm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.self_attn.c_attn.bias", shape: (1152,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.c_attn.q_weight", shape: (1152, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.c_attn.q_scale", shape: (1152, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_weight", shape: (896, 112), dtype: uint32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_scale", shape: (896, 28), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] [2024-06-06 22:37:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.norm.weight", shape: (896,), dtype: float32 + 92%|█████████▏| 157/170 [00:05<00:00, 92.75it/s] 100%|██████████| 170/170 [00:05<00:00, 29.38it/s] +[2024-06-06 22:37:01] INFO huggingface_loader.py:197: Unloading HF weight file: /ssd2/models/Qwen2-0.5B-Instruct/model.safetensors +[2024-06-06 22:37:01] INFO stats.py:77: Time usage: HF loading: 1.540 sec; Pre-quantization mapping: 0.347 sec; Quantization: 2.454 sec +[2024-06-06 22:37:01] INFO stats.py:91: RAM usage: Peak RAM: 1.840 GB. Total bytes loaded from disk: 1.840 GB +[2024-06-06 22:37:01] INFO convert_weight.py:155: Parameter size after quantization: 0.288 GB +[2024-06-06 22:37:01] INFO convert_weight.py:160: Total parameters: 494,032,768 +[2024-06-06 22:37:01] INFO convert_weight.py:161: Bits per parameter: 5.004 +[2024-06-06 22:37:01] INFO convert_weight.py:166: Saved to directory: /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC + +All finished, 8 total shards committed, record saved to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/ndarray-cache.json +Also saved a bf16 record to /models/mlc-delivery/hf/mlc-ai/Qwen2-0.5B-Instruct-q4f32_1-MLC/ndarray-cache-b16.json