Hugging Face's logo

mlc-ai
/
Qwen2-7B-Instruct-q4f32_1-MLC

MLC-LLM
web-llm
Model card Files
xet
Community
Qwen2-7B-Instruct-q4f32_1-MLC / logs.txt
riczhou's picture
riczhou
Initial commit
0e55a09 verified
raw
history blame contribute delete
100 kB
 /opt/conda/envs/py310/bin/python -m mlc_llm gen_config /models/Qwen2-7B-Instruct --quantization q4f32_1 --conv-template chatml --output /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC
[2024-06-06 23:49:11] INFO auto_config.py:116: Found model configuration: /models/Qwen2-7B-Instruct/config.json
[2024-06-06 23:49:11] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override.
[2024-06-06 23:49:11] INFO qwen2_model.py:49: context_window_size not found in config.json. Falling back to max_position_embeddings (32768)
[2024-06-06 23:49:11] INFO qwen2_model.py:66: prefill_chunk_size defaults to 2048
[2024-06-06 23:49:11] INFO config.py:107: Overriding max_batch_size from 1 to 80
[2024-06-06 23:49:11] INFO gen_config.py:143: [generation_config.json] Setting bos_token_id: 151643
[2024-06-06 23:49:11] INFO gen_config.py:143: [generation_config.json] Setting pad_token_id: 151643
[2024-06-06 23:49:11] INFO gen_config.py:143: [generation_config.json] Setting eos_token_id: [151645, 151643]
[2024-06-06 23:49:11] INFO gen_config.py:143: [generation_config.json] Setting repetition_penalty: 1.05
[2024-06-06 23:49:11] INFO gen_config.py:143: [generation_config.json] Setting temperature: 0.7
[2024-06-06 23:49:11] INFO gen_config.py:143: [generation_config.json] Setting top_p: 0.8
[2024-06-06 23:49:11] INFO gen_config.py:157: Not found tokenizer config: /models/Qwen2-7B-Instruct/tokenizer.model
[2024-06-06 23:49:11] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-7B-Instruct/tokenizer.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/tokenizer.json
[2024-06-06 23:49:11] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-7B-Instruct/vocab.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/vocab.json
[2024-06-06 23:49:11] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-7B-Instruct/merges.txt. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/merges.txt
[2024-06-06 23:49:11] INFO gen_config.py:157: Not found tokenizer config: /models/Qwen2-7B-Instruct/added_tokens.json
[2024-06-06 23:49:11] INFO gen_config.py:155: Found tokenizer config: /models/Qwen2-7B-Instruct/tokenizer_config.json. Copying to /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/tokenizer_config.json
[2024-06-06 23:49:12] 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 23:49:12] INFO gen_config.py:32: [System default] Setting presence_penalty: 0.0
[2024-06-06 23:49:12] INFO gen_config.py:32: [System default] Setting frequency_penalty: 0.0
[2024-06-06 23:49:12] INFO gen_config.py:223: Dumping configuration file to: /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/mlc-chat-config.json
/opt/conda/envs/py310/bin/python -m mlc_llm convert_weight /models/Qwen2-7B-Instruct --quantization q4f32_1 --output /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC
[2024-06-06 23:49:13] INFO auto_config.py:116: Found model configuration: /models/Qwen2-7B-Instruct/config.json
[2024-06-06 23:49:14] INFO auto_device.py:79: Found device: cuda:0
[2024-06-06 23:49:16] INFO auto_device.py:88: Not found device: rocm:0
[2024-06-06 23:49:17] INFO auto_device.py:88: Not found device: metal:0
[2024-06-06 23:49:19] INFO auto_device.py:79: Found device: vulkan:0
[2024-06-06 23:49:19] INFO auto_device.py:79: Found device: vulkan:1
[2024-06-06 23:49:19] INFO auto_device.py:79: Found device: vulkan:2
[2024-06-06 23:49:19] INFO auto_device.py:79: Found device: vulkan:3
[2024-06-06 23:49:20] INFO auto_device.py:88: Not found device: opencl:0
[2024-06-06 23:49:20] INFO auto_device.py:35: Using device: cuda:0
[2024-06-06 23:49:20] INFO auto_weight.py:71: Finding weights in: /models/Qwen2-7B-Instruct
[2024-06-06 23:49:20] INFO auto_weight.py:137: Not found Huggingface PyTorch
[2024-06-06 23:49:20] INFO auto_weight.py:144: Found source weight format: huggingface-safetensor. Source configuration: /models/Qwen2-7B-Instruct/model.safetensors.index.json
[2024-06-06 23:49:20] INFO auto_weight.py:107: Using source weight configuration: /models/Qwen2-7B-Instruct/model.safetensors.index.json. Use `--source` to override.
[2024-06-06 23:49:20] INFO auto_weight.py:111: Using source weight format: huggingface-safetensor. Use `--source-format` to override.
[2024-06-06 23:49:20] INFO auto_config.py:154: Found model type: qwen2. Use `--model-type` to override.
[2024-06-06 23:49:20] INFO qwen2_model.py:49: context_window_size not found in config.json. Falling back to max_position_embeddings (32768)
[2024-06-06 23:49:20] INFO qwen2_model.py:66: prefill_chunk_size defaults to 2048
Weight conversion with arguments:
--config /models/Qwen2-7B-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/Qwen2-7B-Instruct/model.safetensors.index.json
--source-format huggingface-safetensor
--output /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC
Start storing to cache /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC
0%| | 0/199 [00:00<?, ?it/s] [2024-06-06 23:49:22] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Qwen2-7B-Instruct/model-00004-of-00004.safetensors
0%| | 0/199 [00:00<?, ?it/s] [2024-06-06 23:49:27] INFO group_quantization.py:217: Compiling quantize function for key: ((152064, 3584), float32, cuda, axis=1, output_transpose=False)
0%| | 0/199 [00:05<?, ?it/s] [2024-06-06 23:49:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "lm_head.q_weight", shape: (152064, 448), dtype: uint32
0%| | 0/199 [00:06<?, ?it/s] [2024-06-06 23:49:29] INFO huggingface_loader.py:167: [Quantized] Parameter: "lm_head.q_scale", shape: (152064, 112), dtype: float32
0%| | 0/199 [00:06<?, ?it/s] 1%| | 1/199 [00:07<23:38, 7.16s/it] [2024-06-06 23:49:29] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.input_layernorm.weight", shape: (3584,), dtype: float32
1%| | 1/199 [00:07<23:38, 7.16s/it] [2024-06-06 23:49:29] INFO group_quantization.py:217: Compiling quantize function for key: ((3584, 18944), float32, cuda, axis=1, output_transpose=False)
1%| | 1/199 [00:07<23:38, 7.16s/it] [2024-06-06 23:49:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
1%| | 1/199 [00:07<23:38, 7.16s/it] [2024-06-06 23:49:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
1%| | 1/199 [00:07<23:38, 7.16s/it] 2%|▏ | 3/199 [00:07<06:49, 2.09s/it] [2024-06-06 23:49:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.post_attention_layernorm.weight", shape: (3584,), dtype: float32
2%|▏ | 3/199 [00:07<06:49, 2.09s/it] [2024-06-06 23:49:30] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.input_layernorm.weight", shape: (3584,), dtype: float32
2%|▏ | 3/199 [00:07<06:49, 2.09s/it] [2024-06-06 23:49:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
2%|▏ | 3/199 [00:07<06:49, 2.09s/it] [2024-06-06 23:49:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
2%|▏ | 3/199 [00:07<06:49, 2.09s/it] 3%|▎ | 6/199 [00:07<02:42, 1.19it/s] [2024-06-06 23:49:30] INFO group_quantization.py:217: Compiling quantize function for key: ((37888, 3584), float32, cuda, axis=1, output_transpose=False)
3%|▎ | 6/199 [00:08<02:42, 1.19it/s] [2024-06-06 23:49:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
3%|▎ | 6/199 [00:08<02:42, 1.19it/s] [2024-06-06 23:49:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
3%|▎ | 6/199 [00:08<02:42, 1.19it/s] 4%|▎ | 7/199 [00:08<02:39, 1.20it/s] [2024-06-06 23:49:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.post_attention_layernorm.weight", shape: (3584,), dtype: float32
4%|▎ | 7/199 [00:08<02:39, 1.20it/s] [2024-06-06 23:49:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.self_attn.c_attn.bias", shape: (4608,), dtype: float32
4%|▎ | 7/199 [00:08<02:39, 1.20it/s] [2024-06-06 23:49:31] INFO group_quantization.py:217: Compiling quantize function for key: ((4608, 3584), float32, cuda, axis=1, output_transpose=False)
4%|▎ | 7/199 [00:08<02:39, 1.20it/s] [2024-06-06 23:49:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
4%|▎ | 7/199 [00:09<02:39, 1.20it/s] [2024-06-06 23:49:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
4%|▎ | 7/199 [00:09<02:39, 1.20it/s] 5%|▌ | 10/199 [00:09<01:34, 2.01it/s] [2024-06-06 23:49:31] INFO group_quantization.py:217: Compiling quantize function for key: ((3584, 3584), float32, cuda, axis=1, output_transpose=False)
5%|▌ | 10/199 [00:09<01:34, 2.01it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
5%|▌ | 10/199 [00:09<01:34, 2.01it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
5%|▌ | 10/199 [00:09<01:34, 2.01it/s] 6%|▌ | 11/199 [00:09<01:31, 2.05it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.24.input_layernorm.weight", shape: (3584,), dtype: float32
6%|▌ | 11/199 [00:09<01:31, 2.05it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
6%|▌ | 11/199 [00:09<01:31, 2.05it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
6%|▌ | 11/199 [00:09<01:31, 2.05it/s] 7%|▋ | 13/199 [00:09<01:06, 2.81it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
7%|▋ | 13/199 [00:10<01:06, 2.81it/s] [2024-06-06 23:49:32] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
7%|▋ | 13/199 [00:10<01:06, 2.81it/s] 7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.24.post_attention_layernorm.weight", shape: (3584,), dtype: float32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.24.self_attn.c_attn.bias", shape: (4608,), dtype: float32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.25.input_layernorm.weight", shape: (3584,), dtype: float32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
7%|▋ | 14/199 [00:10<01:07, 2.76it/s] 10%|█ | 20/199 [00:10<00:27, 6.40it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
10%|█ | 20/199 [00:10<00:27, 6.40it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
10%|█ | 20/199 [00:10<00:27, 6.40it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.25.post_attention_layernorm.weight", shape: (3584,), dtype: float32
10%|█ | 20/199 [00:11<00:27, 6.40it/s] 11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.25.self_attn.c_attn.bias", shape: (4608,), dtype: float32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.26.input_layernorm.weight", shape: (3584,), dtype: float32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] [2024-06-06 23:49:33] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
11%|█ | 22/199 [00:11<00:31, 5.57it/s] 14%|█▎ | 27/199 [00:11<00:20, 8.51it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
14%|█▎ | 27/199 [00:11<00:20, 8.51it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
14%|█▎ | 27/199 [00:11<00:20, 8.51it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.26.post_attention_layernorm.weight", shape: (3584,), dtype: float32
14%|█▎ | 27/199 [00:11<00:20, 8.51it/s] 15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.26.self_attn.c_attn.bias", shape: (4608,), dtype: float32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.27.input_layernorm.weight", shape: (3584,), dtype: float32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
15%|█▍ | 29/199 [00:11<00:23, 7.20it/s] 17%|█▋ | 34/199 [00:11<00:16, 10.23it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
17%|█▋ | 34/199 [00:12<00:16, 10.23it/s] [2024-06-06 23:49:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
17%|█▋ | 34/199 [00:12<00:16, 10.23it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.27.post_attention_layernorm.weight", shape: (3584,), dtype: float32
17%|█▋ | 34/199 [00:12<00:16, 10.23it/s] 18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.27.self_attn.c_attn.bias", shape: (4608,), dtype: float32
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.norm.weight", shape: (3584,), dtype: float32
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Qwen2-7B-Instruct/model-00004-of-00004.safetensors
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:35] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Qwen2-7B-Instruct/model-00001-of-00004.safetensors
18%|█▊ | 36/199 [00:12<00:19, 8.20it/s] [2024-06-06 23:49:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.embed_tokens.q_weight", shape: (152064, 448), dtype: uint32
18%|█▊ | 36/199 [00:16<00:19, 8.20it/s] [2024-06-06 23:49:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.embed_tokens.q_scale", shape: (152064, 112), dtype: float32
18%|█▊ | 36/199 [00:17<00:19, 8.20it/s] 21%|██ | 41/199 [00:17<01:21, 1.93it/s] [2024-06-06 23:49:40] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.input_layernorm.weight", shape: (3584,), dtype: float32
21%|██ | 41/199 [00:17<01:21, 1.93it/s] [2024-06-06 23:49:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
21%|██ | 41/199 [00:17<01:21, 1.93it/s] [2024-06-06 23:49:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
21%|██ | 41/199 [00:17<01:21, 1.93it/s] 22%|██▏ | 43/199 [00:17<01:07, 2.30it/s] [2024-06-06 23:49:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
22%|██▏ | 43/199 [00:18<01:07, 2.30it/s] [2024-06-06 23:49:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
22%|██▏ | 43/199 [00:18<01:07, 2.30it/s] 22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.post_attention_layernorm.weight", shape: (3584,), dtype: float32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.self_attn.c_attn.bias", shape: (4608,), dtype: float32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.input_layernorm.weight", shape: (3584,), dtype: float32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
22%|██▏ | 44/199 [00:18<01:07, 2.31it/s] 25%|██▌ | 50/199 [00:18<00:34, 4.34it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
25%|██▌ | 50/199 [00:18<00:34, 4.34it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
25%|██▌ | 50/199 [00:18<00:34, 4.34it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.post_attention_layernorm.weight", shape: (3584,), dtype: float32
25%|██▌ | 50/199 [00:18<00:34, 4.34it/s] 26%|██▌ | 52/199 [00:18<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.self_attn.c_attn.bias", shape: (4608,), dtype: float32
26%|██▌ | 52/199 [00:18<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
26%|██▌ | 52/199 [00:18<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
26%|██▌ | 52/199 [00:18<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
26%|██▌ | 52/199 [00:19<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
26%|██▌ | 52/199 [00:19<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.input_layernorm.weight", shape: (3584,), dtype: float32
26%|██▌ | 52/199 [00:19<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
26%|██▌ | 52/199 [00:19<00:33, 4.40it/s] [2024-06-06 23:49:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
26%|██▌ | 52/199 [00:19<00:33, 4.40it/s] 29%|██▊ | 57/199 [00:19<00:21, 6.60it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
29%|██▊ | 57/199 [00:19<00:21, 6.60it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
29%|██▊ | 57/199 [00:19<00:21, 6.60it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.post_attention_layernorm.weight", shape: (3584,), dtype: float32
29%|██▊ | 57/199 [00:19<00:21, 6.60it/s] 30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.self_attn.c_attn.bias", shape: (4608,), dtype: float32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.input_layernorm.weight", shape: (3584,), dtype: float32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
30%|██▉ | 59/199 [00:19<00:23, 6.08it/s] 32%|███▏ | 64/199 [00:19<00:15, 8.70it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
32%|███▏ | 64/199 [00:20<00:15, 8.70it/s] [2024-06-06 23:49:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
32%|███▏ | 64/199 [00:20<00:15, 8.70it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.post_attention_layernorm.weight", shape: (3584,), dtype: float32
32%|███▏ | 64/199 [00:20<00:15, 8.70it/s] 33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.self_attn.c_attn.bias", shape: (4608,), dtype: float32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.input_layernorm.weight", shape: (3584,), dtype: float32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
33%|███▎ | 66/199 [00:20<00:18, 7.39it/s] 36%|███▌ | 71/199 [00:20<00:12, 10.19it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
36%|███▌ | 71/199 [00:20<00:12, 10.19it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
36%|███▌ | 71/199 [00:20<00:12, 10.19it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.post_attention_layernorm.weight", shape: (3584,), dtype: float32
36%|███▌ | 71/199 [00:20<00:12, 10.19it/s] 37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.self_attn.c_attn.bias", shape: (4608,), dtype: float32
37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.input_layernorm.weight", shape: (3584,), dtype: float32
37%|███▋ | 73/199 [00:20<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
37%|███▋ | 73/199 [00:21<00:15, 8.32it/s] [2024-06-06 23:49:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
37%|███▋ | 73/199 [00:21<00:15, 8.32it/s] 39%|███▉ | 78/199 [00:21<00:10, 11.28it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
39%|███▉ | 78/199 [00:21<00:10, 11.28it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
39%|███▉ | 78/199 [00:21<00:10, 11.28it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.post_attention_layernorm.weight", shape: (3584,), dtype: float32
39%|███▉ | 78/199 [00:21<00:10, 11.28it/s] 40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.self_attn.c_attn.bias", shape: (4608,), dtype: float32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.self_attn.c_attn.bias", shape: (4608,), dtype: float32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
40%|████ | 80/199 [00:21<00:13, 8.93it/s] 43%|████▎ | 85/199 [00:21<00:08, 12.69it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
43%|████▎ | 85/199 [00:21<00:08, 12.69it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
43%|████▎ | 85/199 [00:21<00:08, 12.69it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Qwen2-7B-Instruct/model-00001-of-00004.safetensors
43%|████▎ | 85/199 [00:21<00:08, 12.69it/s] [2024-06-06 23:49:44] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Qwen2-7B-Instruct/model-00002-of-00004.safetensors
43%|████▎ | 85/199 [00:22<00:08, 12.69it/s] [2024-06-06 23:49:47] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.input_layernorm.weight", shape: (3584,), dtype: float32
43%|████▎ | 85/199 [00:25<00:08, 12.69it/s] 44%|████▎ | 87/199 [00:25<00:43, 2.57it/s] [2024-06-06 23:49:47] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
44%|████▎ | 87/199 [00:25<00:43, 2.57it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
44%|████▎ | 87/199 [00:25<00:43, 2.57it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
44%|████▎ | 87/199 [00:25<00:43, 2.57it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
44%|████▎ | 87/199 [00:25<00:43, 2.57it/s] 45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.post_attention_layernorm.weight", shape: (3584,), dtype: float32
45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.self_attn.c_attn.bias", shape: (4608,), dtype: float32
45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
45%|████▍ | 89/199 [00:25<00:40, 2.73it/s] 47%|████▋ | 93/199 [00:25<00:25, 4.18it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.input_layernorm.weight", shape: (3584,), dtype: float32
47%|████▋ | 93/199 [00:25<00:25, 4.18it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
47%|████▋ | 93/199 [00:25<00:25, 4.18it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
47%|████▋ | 93/199 [00:26<00:25, 4.18it/s] 48%|████▊ | 95/199 [00:26<00:21, 4.87it/s] [2024-06-06 23:49:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
48%|████▊ | 95/199 [00:26<00:21, 4.87it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
48%|████▊ | 95/199 [00:26<00:21, 4.87it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.post_attention_layernorm.weight", shape: (3584,), dtype: float32
48%|████▊ | 95/199 [00:26<00:21, 4.87it/s] 49%|████▊ | 97/199 [00:26<00:20, 4.97it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.self_attn.c_attn.bias", shape: (4608,), dtype: float32
49%|████▊ | 97/199 [00:26<00:20, 4.97it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
49%|████▊ | 97/199 [00:26<00:20, 4.97it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
49%|████▊ | 97/199 [00:26<00:20, 4.97it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
49%|████▊ | 97/199 [00:26<00:20, 4.97it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
49%|████▊ | 97/199 [00:26<00:20, 4.97it/s] 50%|█████ | 100/199 [00:26<00:14, 6.85it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.input_layernorm.weight", shape: (3584,), dtype: float32
50%|█████ | 100/199 [00:26<00:14, 6.85it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
50%|█████ | 100/199 [00:26<00:14, 6.85it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
50%|█████ | 100/199 [00:26<00:14, 6.85it/s] 51%|█████▏ | 102/199 [00:26<00:12, 7.70it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
51%|█████▏ | 102/199 [00:26<00:12, 7.70it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
51%|█████▏ | 102/199 [00:27<00:12, 7.70it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.post_attention_layernorm.weight", shape: (3584,), dtype: float32
51%|█████▏ | 102/199 [00:27<00:12, 7.70it/s] 52%|█████▏ | 104/199 [00:27<00:14, 6.72it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.self_attn.c_attn.bias", shape: (4608,), dtype: float32
52%|█████▏ | 104/199 [00:27<00:14, 6.72it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
52%|█████▏ | 104/199 [00:27<00:14, 6.72it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
52%|█████▏ | 104/199 [00:27<00:14, 6.72it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
52%|█████▏ | 104/199 [00:27<00:14, 6.72it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
52%|█████▏ | 104/199 [00:27<00:14, 6.72it/s] 54%|█████▍ | 107/199 [00:27<00:09, 9.21it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.input_layernorm.weight", shape: (3584,), dtype: float32
54%|█████▍ | 107/199 [00:27<00:09, 9.21it/s] [2024-06-06 23:49:49] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
54%|█████▍ | 107/199 [00:27<00:09, 9.21it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
54%|█████▍ | 107/199 [00:27<00:09, 9.21it/s] 55%|█████▍ | 109/199 [00:27<00:09, 9.76it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
55%|█████▍ | 109/199 [00:27<00:09, 9.76it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
55%|█████▍ | 109/199 [00:27<00:09, 9.76it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.post_attention_layernorm.weight", shape: (3584,), dtype: float32
55%|█████▍ | 109/199 [00:27<00:09, 9.76it/s] 56%|█████▌ | 111/199 [00:27<00:11, 7.80it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.self_attn.c_attn.bias", shape: (4608,), dtype: float32
56%|█████▌ | 111/199 [00:27<00:11, 7.80it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
56%|█████▌ | 111/199 [00:27<00:11, 7.80it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
56%|█████▌ | 111/199 [00:27<00:11, 7.80it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
56%|█████▌ | 111/199 [00:27<00:11, 7.80it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
56%|█████▌ | 111/199 [00:27<00:11, 7.80it/s] 57%|█████▋ | 114/199 [00:27<00:08, 10.55it/s] [2024-06-06 23:49:50] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Qwen2-7B-Instruct/model-00003-of-00004.safetensors
57%|█████▋ | 114/199 [00:27<00:08, 10.55it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
57%|█████▋ | 114/199 [00:32<00:08, 10.55it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
57%|█████▋ | 114/199 [00:32<00:08, 10.55it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.self_attn.c_attn.bias", shape: (4608,), dtype: float32
57%|█████▋ | 114/199 [00:32<00:08, 10.55it/s] 58%|█████▊ | 116/199 [00:32<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
58%|█████▊ | 116/199 [00:32<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
58%|█████▊ | 116/199 [00:32<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
58%|█████▊ | 116/199 [00:32<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
58%|█████▊ | 116/199 [00:32<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.input_layernorm.weight", shape: (3584,), dtype: float32
58%|█████▊ | 116/199 [00:32<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
58%|█████▊ | 116/199 [00:33<00:58, 1.41it/s] [2024-06-06 23:49:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
58%|█████▊ | 116/199 [00:33<00:58, 1.41it/s] 60%|██████ | 120/199 [00:33<00:35, 2.21it/s] [2024-06-06 23:49:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
60%|██████ | 120/199 [00:34<00:35, 2.21it/s] [2024-06-06 23:49:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
60%|██████ | 120/199 [00:34<00:35, 2.21it/s] [2024-06-06 23:49:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.post_attention_layernorm.weight", shape: (3584,), dtype: float32
60%|██████ | 120/199 [00:34<00:35, 2.21it/s] 61%|██████▏ | 122/199 [00:34<00:39, 1.97it/s] [2024-06-06 23:49:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.input_layernorm.weight", shape: (3584,), dtype: float32
61%|██████▏ | 122/199 [00:34<00:39, 1.97it/s] [2024-06-06 23:49:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
61%|██████▏ | 122/199 [00:34<00:39, 1.97it/s] [2024-06-06 23:49:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
61%|██████▏ | 122/199 [00:34<00:39, 1.97it/s] 62%|██████▏ | 124/199 [00:34<00:30, 2.49it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
62%|██████▏ | 124/199 [00:35<00:30, 2.49it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
62%|██████▏ | 124/199 [00:35<00:30, 2.49it/s] 63%|██████▎ | 125/199 [00:35<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.post_attention_layernorm.weight", shape: (3584,), dtype: float32
63%|██████▎ | 125/199 [00:35<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.self_attn.c_attn.bias", shape: (4608,), dtype: float32
63%|██████▎ | 125/199 [00:35<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
63%|██████▎ | 125/199 [00:35<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
63%|██████▎ | 125/199 [00:36<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
63%|██████▎ | 125/199 [00:36<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
63%|██████▎ | 125/199 [00:36<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.input_layernorm.weight", shape: (3584,), dtype: float32
63%|██████▎ | 125/199 [00:36<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
63%|██████▎ | 125/199 [00:36<00:38, 1.95it/s] [2024-06-06 23:49:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
63%|██████▎ | 125/199 [00:36<00:38, 1.95it/s] 66%|██████▌ | 131/199 [00:36<00:16, 4.02it/s] [2024-06-06 23:49:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
66%|██████▌ | 131/199 [00:36<00:16, 4.02it/s] [2024-06-06 23:49:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
66%|██████▌ | 131/199 [00:37<00:16, 4.02it/s] [2024-06-06 23:49:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.post_attention_layernorm.weight", shape: (3584,), dtype: float32
66%|██████▌ | 131/199 [00:37<00:16, 4.02it/s] 67%|██████▋ | 133/199 [00:37<00:20, 3.28it/s] [2024-06-06 23:49:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.self_attn.c_attn.bias", shape: (4608,), dtype: float32
67%|██████▋ | 133/199 [00:37<00:20, 3.28it/s] [2024-06-06 23:49:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
67%|██████▋ | 133/199 [00:37<00:20, 3.28it/s] [2024-06-06 23:50:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
67%|██████▋ | 133/199 [00:37<00:20, 3.28it/s] [2024-06-06 23:50:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
67%|██████▋ | 133/199 [00:37<00:20, 3.28it/s] [2024-06-06 23:50:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
67%|██████▋ | 133/199 [00:37<00:20, 3.28it/s] 68%|██████▊ | 136/199 [00:37<00:13, 4.53it/s] [2024-06-06 23:50:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.input_layernorm.weight", shape: (3584,), dtype: float32
68%|██████▊ | 136/199 [00:37<00:13, 4.53it/s] [2024-06-06 23:50:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
68%|██████▊ | 136/199 [00:37<00:13, 4.53it/s] [2024-06-06 23:50:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
68%|██████▊ | 136/199 [00:37<00:13, 4.53it/s] 69%|██████▉ | 138/199 [00:37<00:11, 5.28it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
69%|██████▉ | 138/199 [00:38<00:11, 5.28it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
69%|██████▉ | 138/199 [00:38<00:11, 5.28it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.post_attention_layernorm.weight", shape: (3584,), dtype: float32
69%|██████▉ | 138/199 [00:38<00:11, 5.28it/s] 70%|███████ | 140/199 [00:38<00:17, 3.46it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.self_attn.c_attn.bias", shape: (4608,), dtype: float32
70%|███████ | 140/199 [00:38<00:17, 3.46it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
70%|███████ | 140/199 [00:38<00:17, 3.46it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
70%|███████ | 140/199 [00:38<00:17, 3.46it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
70%|███████ | 140/199 [00:38<00:17, 3.46it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
70%|███████ | 140/199 [00:38<00:17, 3.46it/s] 72%|███████▏ | 143/199 [00:38<00:11, 4.97it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.input_layernorm.weight", shape: (3584,), dtype: float32
72%|███████▏ | 143/199 [00:38<00:11, 4.97it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
72%|███████▏ | 143/199 [00:38<00:11, 4.97it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
72%|███████▏ | 143/199 [00:38<00:11, 4.97it/s] 73%|███████▎ | 145/199 [00:38<00:09, 5.82it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.post_attention_layernorm.weight", shape: (3584,), dtype: float32
73%|███████▎ | 145/199 [00:38<00:09, 5.82it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.input_layernorm.weight", shape: (3584,), dtype: float32
73%|███████▎ | 145/199 [00:38<00:09, 5.82it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
73%|███████▎ | 145/199 [00:39<00:09, 5.82it/s] [2024-06-06 23:50:01] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
73%|███████▎ | 145/199 [00:39<00:09, 5.82it/s] 74%|███████▍ | 148/199 [00:39<00:06, 7.61it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
74%|███████▍ | 148/199 [00:39<00:06, 7.61it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
74%|███████▍ | 148/199 [00:40<00:06, 7.61it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.post_attention_layernorm.weight", shape: (3584,), dtype: float32
74%|███████▍ | 148/199 [00:40<00:06, 7.61it/s] 75%|███████▌ | 150/199 [00:40<00:10, 4.48it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.self_attn.c_attn.bias", shape: (4608,), dtype: float32
75%|███████▌ | 150/199 [00:40<00:10, 4.48it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
75%|███████▌ | 150/199 [00:40<00:10, 4.48it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
75%|███████▌ | 150/199 [00:40<00:10, 4.48it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
75%|███████▌ | 150/199 [00:40<00:10, 4.48it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
75%|███████▌ | 150/199 [00:40<00:10, 4.48it/s] 77%|███████▋ | 153/199 [00:40<00:07, 6.30it/s] [2024-06-06 23:50:02] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.input_layernorm.weight", shape: (3584,), dtype: float32
77%|███████▋ | 153/199 [00:40<00:07, 6.30it/s] [2024-06-06 23:50:03] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
77%|███████▋ | 153/199 [00:40<00:07, 6.30it/s] [2024-06-06 23:50:03] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
77%|███████▋ | 153/199 [00:40<00:07, 6.30it/s] 78%|███████▊ | 155/199 [00:40<00:06, 7.15it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
78%|███████▊ | 155/199 [00:41<00:06, 7.15it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
78%|███████▊ | 155/199 [00:41<00:06, 7.15it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.post_attention_layernorm.weight", shape: (3584,), dtype: float32
78%|███████▊ | 155/199 [00:41<00:06, 7.15it/s] 79%|███████▉ | 157/199 [00:41<00:10, 4.06it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.self_attn.c_attn.bias", shape: (4608,), dtype: float32
79%|███████▉ | 157/199 [00:41<00:10, 4.06it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
79%|███████▉ | 157/199 [00:41<00:10, 4.06it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
79%|███████▉ | 157/199 [00:41<00:10, 4.06it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
79%|███████▉ | 157/199 [00:41<00:10, 4.06it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
79%|███████▉ | 157/199 [00:41<00:10, 4.06it/s] 80%|████████ | 160/199 [00:41<00:06, 5.87it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.input_layernorm.weight", shape: (3584,), dtype: float32
80%|████████ | 160/199 [00:41<00:06, 5.87it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
80%|████████ | 160/199 [00:41<00:06, 5.87it/s] [2024-06-06 23:50:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
80%|████████ | 160/199 [00:41<00:06, 5.87it/s] 81%|████████▏ | 162/199 [00:41<00:05, 6.75it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
81%|████████▏ | 162/199 [00:42<00:05, 6.75it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
81%|████████▏ | 162/199 [00:42<00:05, 6.75it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.post_attention_layernorm.weight", shape: (3584,), dtype: float32
81%|████████▏ | 162/199 [00:42<00:05, 6.75it/s] 82%|████████▏ | 164/199 [00:42<00:09, 3.73it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.self_attn.c_attn.bias", shape: (4608,), dtype: float32
82%|████████▏ | 164/199 [00:42<00:09, 3.73it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
82%|████████▏ | 164/199 [00:42<00:09, 3.73it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
82%|████████▏ | 164/199 [00:43<00:09, 3.73it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
82%|████████▏ | 164/199 [00:43<00:09, 3.73it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
82%|████████▏ | 164/199 [00:43<00:09, 3.73it/s] 84%|████████▍ | 167/199 [00:43<00:05, 5.45it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.input_layernorm.weight", shape: (3584,), dtype: float32
84%|████████▍ | 167/199 [00:43<00:05, 5.45it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
84%|████████▍ | 167/199 [00:43<00:05, 5.45it/s] [2024-06-06 23:50:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
84%|████████▍ | 167/199 [00:43<00:05, 5.45it/s] 85%|████████▍ | 169/199 [00:43<00:04, 6.34it/s] [2024-06-06 23:50:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
85%|████████▍ | 169/199 [00:43<00:04, 6.34it/s] [2024-06-06 23:50:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
85%|████████▍ | 169/199 [00:44<00:04, 6.34it/s] [2024-06-06 23:50:06] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.post_attention_layernorm.weight", shape: (3584,), dtype: float32
85%|████████▍ | 169/199 [00:44<00:04, 6.34it/s] 86%|████████▌ | 171/199 [00:44<00:06, 4.04it/s] [2024-06-06 23:50:06] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.self_attn.c_attn.bias", shape: (4608,), dtype: float32
86%|████████▌ | 171/199 [00:44<00:06, 4.04it/s] [2024-06-06 23:50:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
86%|████████▌ | 171/199 [00:44<00:06, 4.04it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
86%|████████▌ | 171/199 [00:44<00:06, 4.04it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
86%|████████▌ | 171/199 [00:44<00:06, 4.04it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
86%|████████▌ | 171/199 [00:44<00:06, 4.04it/s] 87%|████████▋ | 174/199 [00:44<00:04, 5.89it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.input_layernorm.weight", shape: (3584,), dtype: float32
87%|████████▋ | 174/199 [00:44<00:04, 5.89it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
87%|████████▋ | 174/199 [00:44<00:04, 5.89it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
87%|████████▋ | 174/199 [00:44<00:04, 5.89it/s] 88%|████████▊ | 176/199 [00:44<00:03, 6.79it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
88%|████████▊ | 176/199 [00:45<00:03, 6.79it/s] [2024-06-06 23:50:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
88%|████████▊ | 176/199 [00:45<00:03, 6.79it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.post_attention_layernorm.weight", shape: (3584,), dtype: float32
88%|████████▊ | 176/199 [00:45<00:03, 6.79it/s] 89%|████████▉ | 178/199 [00:45<00:04, 4.62it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.self_attn.c_attn.bias", shape: (4608,), dtype: float32
89%|████████▉ | 178/199 [00:45<00:04, 4.62it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
89%|████████▉ | 178/199 [00:45<00:04, 4.62it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
89%|████████▉ | 178/199 [00:45<00:04, 4.62it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
89%|████████▉ | 178/199 [00:45<00:04, 4.62it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
89%|████████▉ | 178/199 [00:45<00:04, 4.62it/s] 91%|█████████ | 181/199 [00:45<00:02, 6.67it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.input_layernorm.weight", shape: (3584,), dtype: float32
91%|█████████ | 181/199 [00:45<00:02, 6.67it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
91%|█████████ | 181/199 [00:45<00:02, 6.67it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
91%|█████████ | 181/199 [00:45<00:02, 6.67it/s] 92%|█████████▏| 183/199 [00:45<00:02, 7.54it/s] [2024-06-06 23:50:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
92%|█████████▏| 183/199 [00:46<00:02, 7.54it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
92%|█████████▏| 183/199 [00:46<00:02, 7.54it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.post_attention_layernorm.weight", shape: (3584,), dtype: float32
92%|█████████▏| 183/199 [00:46<00:02, 7.54it/s] 93%|█████████▎| 185/199 [00:46<00:02, 4.72it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.self_attn.c_attn.bias", shape: (4608,), dtype: float32
93%|█████████▎| 185/199 [00:46<00:02, 4.72it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
93%|█████████▎| 185/199 [00:46<00:02, 4.72it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
93%|█████████▎| 185/199 [00:46<00:02, 4.72it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
93%|█████████▎| 185/199 [00:46<00:02, 4.72it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
93%|█████████▎| 185/199 [00:46<00:02, 4.72it/s] 94%|█████████▍| 188/199 [00:46<00:01, 6.79it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.input_layernorm.weight", shape: (3584,), dtype: float32
94%|█████████▍| 188/199 [00:46<00:01, 6.79it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_weight", shape: (3584, 2368), dtype: uint32
94%|█████████▍| 188/199 [00:46<00:01, 6.79it/s] [2024-06-06 23:50:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_scale", shape: (3584, 592), dtype: float32
94%|█████████▍| 188/199 [00:46<00:01, 6.79it/s] 95%|█████████▌| 190/199 [00:46<00:01, 7.68it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
95%|█████████▌| 190/199 [00:47<00:01, 7.68it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
95%|█████████▌| 190/199 [00:47<00:01, 7.68it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.post_attention_layernorm.weight", shape: (3584,), dtype: float32
95%|█████████▌| 190/199 [00:47<00:01, 7.68it/s] 96%|█████████▋| 192/199 [00:47<00:01, 4.71it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.self_attn.c_attn.bias", shape: (4608,), dtype: float32
96%|█████████▋| 192/199 [00:47<00:01, 4.71it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
96%|█████████▋| 192/199 [00:47<00:01, 4.71it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
96%|█████████▋| 192/199 [00:47<00:01, 4.71it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
96%|█████████▋| 192/199 [00:47<00:01, 4.71it/s] [2024-06-06 23:50:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
96%|█████████▋| 192/199 [00:47<00:01, 4.71it/s] 98%|█████████▊| 195/199 [00:47<00:00, 6.79it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.gate_up_proj.q_weight", shape: (37888, 448), dtype: uint32
98%|█████████▊| 195/199 [00:48<00:00, 6.79it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.mlp.gate_up_proj.q_scale", shape: (37888, 112), dtype: float32
98%|█████████▊| 195/199 [00:48<00:00, 6.79it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.self_attn.c_attn.bias", shape: (4608,), dtype: float32
98%|█████████▊| 195/199 [00:48<00:00, 6.79it/s] 99%|█████████▉| 197/199 [00:48<00:00, 4.28it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.c_attn.q_weight", shape: (4608, 448), dtype: uint32
99%|█████████▉| 197/199 [00:48<00:00, 4.28it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.c_attn.q_scale", shape: (4608, 112), dtype: float32
99%|█████████▉| 197/199 [00:48<00:00, 4.28it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_weight", shape: (3584, 448), dtype: uint32
99%|█████████▉| 197/199 [00:48<00:00, 4.28it/s] [2024-06-06 23:50:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_scale", shape: (3584, 112), dtype: float32
99%|█████████▉| 197/199 [00:48<00:00, 4.28it/s] 100%|██████████| 199/199 [00:48<00:00, 5.36it/s] 100%|██████████| 199/199 [00:48<00:00, 4.08it/s]
[2024-06-06 23:50:11] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Qwen2-7B-Instruct/model-00002-of-00004.safetensors
[2024-06-06 23:50:11] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Qwen2-7B-Instruct/model-00003-of-00004.safetensors
[2024-06-06 23:50:12] INFO stats.py:77: Time usage: HF loading: 15.376 sec; Pre-quantization mapping: 15.644 sec; Quantization: 3.197 sec
[2024-06-06 23:50:12] INFO stats.py:91: RAM usage: Peak RAM: 14.397 GB. Total bytes loaded from disk: 28.370 GB
[2024-06-06 23:50:12] INFO convert_weight.py:155: Parameter size after quantization: 4.434 GB
[2024-06-06 23:50:12] INFO convert_weight.py:160: Total parameters: 7,615,616,512
[2024-06-06 23:50:12] INFO convert_weight.py:161: Bits per parameter: 5.001
[2024-06-06 23:50:12] INFO convert_weight.py:166: Saved to directory: /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC
All finished, 88 total shards committed, record saved to /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/ndarray-cache.json
Also saved a bf16 record to /models/mlc-delivery/hf/mlc-ai/Qwen2-7B-Instruct-q4f32_1-MLC/ndarray-cache-b16.json