diff --git "a/logs.txt" "b/logs.txt" new file mode 100644--- /dev/null +++ "b/logs.txt" @@ -0,0 +1,314 @@ +/home/floriadmin/miniforge3/envs/mlc/bin/python -m mlc_llm gen_config /tmp/tmpo8_dluj_/repo --quantization q4f32_1 --conv-template chatml --output /tmp/tmpx3x6wnfw +[2024-03-19 01:27:45] INFO auto_config.py:115: Found model configuration: /tmp/tmpo8_dluj_/repo/config.json +[2024-03-19 01:27:45] INFO auto_config.py:153: Found model type: llama. Use `--model-type` to override. +[2024-03-19 01:27:45] INFO llama_model.py:52: context_window_size not found in config.json. Falling back to max_position_embeddings (4096) +[2024-03-19 01:27:45] INFO llama_model.py:72: prefill_chunk_size defaults to context_window_size (4096) +[2024-03-19 01:27:45] INFO config.py:106: Overriding max_batch_size from 1 to 80 +[2024-03-19 01:27:45] INFO gen_config.py:133: [generation_config.json] Setting bos_token_id: 1 +[2024-03-19 01:27:45] INFO gen_config.py:133: [generation_config.json] Setting eos_token_id: 2 +[2024-03-19 01:27:45] INFO gen_config.py:133: [generation_config.json] Setting pad_token_id: 0 +[2024-03-19 01:27:45] INFO gen_config.py:145: Found tokenizer config: /tmp/tmpo8_dluj_/repo/tokenizer.model. Copying to /tmp/tmpx3x6wnfw/tokenizer.model +[2024-03-19 01:27:45] INFO gen_config.py:145: Found tokenizer config: /tmp/tmpo8_dluj_/repo/tokenizer.json. Copying to /tmp/tmpx3x6wnfw/tokenizer.json +[2024-03-19 01:27:45] INFO gen_config.py:147: Not found tokenizer config: /tmp/tmpo8_dluj_/repo/vocab.json +[2024-03-19 01:27:45] INFO gen_config.py:147: Not found tokenizer config: /tmp/tmpo8_dluj_/repo/merges.txt +[2024-03-19 01:27:45] INFO gen_config.py:145: Found tokenizer config: /tmp/tmpo8_dluj_/repo/added_tokens.json. Copying to /tmp/tmpx3x6wnfw/added_tokens.json +[2024-03-19 01:27:45] INFO gen_config.py:145: Found tokenizer config: /tmp/tmpo8_dluj_/repo/tokenizer_config.json. Copying to /tmp/tmpx3x6wnfw/tokenizer_config.json +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting temperature: 0.7 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting presence_penalty: 0.0 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting frequency_penalty: 0.0 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting repetition_penalty: 1.0 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting top_p: 0.95 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting mean_gen_len: 128 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting max_gen_len: 512 +[2024-03-19 01:27:45] INFO gen_config.py:75: [System default] Setting shift_fill_factor: 0.3 +[2024-03-19 01:27:45] INFO gen_config.py:198: Dumping configuration file to: /tmp/tmpx3x6wnfw/mlc-chat-config.json +/home/floriadmin/miniforge3/envs/mlc/bin/python -m mlc_llm convert_weight /tmp/tmpo8_dluj_/repo --quantization q4f32_1 --source-format auto --output /tmp/tmpx3x6wnfw +[2024-03-19 01:27:46] INFO auto_config.py:115: Found model configuration: /tmp/tmpo8_dluj_/repo/config.json +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:0 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:1 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:2 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:3 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:4 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:5 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:6 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:7 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:8 +[2024-03-19 01:27:47] INFO auto_device.py:76: Found device: cuda:9 +[2024-03-19 01:27:48] INFO auto_device.py:85: Not found device: rocm:0 +[2024-03-19 01:27:49] INFO auto_device.py:85: Not found device: metal:0 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:0 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:1 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:2 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:3 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:4 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:5 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:6 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:7 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:8 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:9 +[2024-03-19 01:27:52] INFO auto_device.py:76: Found device: vulkan:10 +[2024-03-19 01:27:53] INFO auto_device.py:85: Not found device: opencl:0 +[2024-03-19 01:27:53] INFO auto_device.py:33: Using device: cuda:0 +[2024-03-19 01:27:53] INFO auto_weight.py:70: Finding weights in: /tmp/tmpo8_dluj_/repo +[2024-03-19 01:27:53] INFO auto_weight.py:136: Not found Huggingface PyTorch +[2024-03-19 01:27:54] INFO auto_weight.py:160: Found source weight format: huggingface-safetensor. Source configuration: /tmp/tmpo8_dluj_/repo/model.safetensors.index.json +[2024-03-19 01:27:54] INFO auto_weight.py:106: Using source weight configuration: /tmp/tmpo8_dluj_/repo/model.safetensors.index.json. Use `--source` to override. +[2024-03-19 01:27:54] INFO auto_weight.py:110: Using source weight format: huggingface-safetensor. Use `--source-format` to override. +[2024-03-19 01:27:54] INFO auto_config.py:153: Found model type: llama. Use `--model-type` to override. +[2024-03-19 01:27:54] INFO llama_model.py:52: context_window_size not found in config.json. Falling back to max_position_embeddings (4096) +[2024-03-19 01:27:54] INFO llama_model.py:72: prefill_chunk_size defaults to context_window_size (4096) +Weight conversion with arguments: + --config /tmp/tmpo8_dluj_/repo/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 llama + --device cuda:0 + --source /tmp/tmpo8_dluj_/repo/model.safetensors.index.json + --source-format huggingface-safetensor + --output /tmp/tmpx3x6wnfw +Start storing to cache /tmp/tmpx3x6wnfw + 0%| | 0/135 [00:00 type is zero. + setattr(self, word, getattr(machar, word).flat[0]) +/home/floriadmin/miniforge3/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%|▌ | 1/135 [00:13<30:04, 13.46s/it] [2024-03-19 01:28:10] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.embed_tokens.q_weight", shape: (32002, 256), dtype: uint32 + 1%|▌ | 1/135 [00:14<30:04, 13.46s/it] [2024-03-19 01:28:10] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.embed_tokens.q_scale", shape: (32002, 64), dtype: float32 + 1%|▌ | 1/135 [00:14<30:04, 13.46s/it] 1%|█▏ | 2/135 [00:14<14:17, 6.45s/it] [2024-03-19 01:28:10] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.0.input_layernorm.weight", shape: (2048,), dtype: float32 + 1%|█▏ | 2/135 [00:14<14:17, 6.45s/it] [2024-03-19 01:28:10] INFO group_quantization.py:232: Compiling quantize function for key: ((2048, 5632), float32, cuda, axis=1, output_transpose=False) + 1%|█▏ | 2/135 [00:15<14:17, 6.45s/it] [2024-03-19 01:28:11] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 1%|█▏ | 2/135 [00:15<14:17, 6.45s/it] [2024-03-19 01:28:11] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 1%|█▏ | 2/135 [00:15<14:17, 6.45s/it] 3%|██▎ | 4/135 [00:15<05:44, 2.63s/it] [2024-03-19 01:28:11] INFO group_quantization.py:232: Compiling quantize function for key: ((11264, 2048), float32, cuda, axis=1, output_transpose=False) + 3%|██▎ | 4/135 [00:16<05:44, 2.63s/it] [2024-03-19 01:28:12] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 3%|██▎ | 4/135 [00:16<05:44, 2.63s/it] [2024-03-19 01:28:12] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 3%|██▎ | 4/135 [00:16<05:44, 2.63s/it] 4%|██▉ | 5/135 [00:16<04:34, 2.11s/it] [2024-03-19 01:28:12] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.0.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 4%|██▉ | 5/135 [00:16<04:34, 2.11s/it] [2024-03-19 01:28:12] INFO group_quantization.py:232: Compiling quantize function for key: ((2560, 2048), float32, cuda, axis=1, output_transpose=False) + 4%|██▉ | 5/135 [00:16<04:34, 2.11s/it] [2024-03-19 01:28:12] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 4%|██▉ | 5/135 [00:17<04:34, 2.11s/it] [2024-03-19 01:28:12] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 4%|██▉ | 5/135 [00:17<04:34, 2.11s/it] 5%|████▏ | 7/135 [00:17<02:38, 1.24s/it] [2024-03-19 01:28:12] INFO group_quantization.py:232: Compiling quantize function for key: ((2048, 2048), float32, cuda, axis=1, output_transpose=False) + 5%|████▏ | 7/135 [00:17<02:38, 1.24s/it] [2024-03-19 01:28:13] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 5%|████▏ | 7/135 [00:17<02:38, 1.24s/it] [2024-03-19 01:28:13] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 5%|████▏ | 7/135 [00:17<02:38, 1.24s/it] 6%|████▋ | 8/135 [00:17<02:13, 1.05s/it] [2024-03-19 01:28:13] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.1.input_layernorm.weight", shape: (2048,), dtype: float32 + 6%|████▋ | 8/135 [00:17<02:13, 1.05s/it] [2024-03-19 01:28:13] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 6%|████▋ | 8/135 [00:17<02:13, 1.05s/it] [2024-03-19 01:28:13] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 6%|████▋ | 8/135 [00:17<02:13, 1.05s/it] 7%|█████▊ | 10/135 [00:17<01:22, 1.51it/s] [2024-03-19 01:28:13] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 7%|█████▊ | 10/135 [00:18<01:22, 1.51it/s] [2024-03-19 01:28:13] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 7%|█████▊ | 10/135 [00:18<01:22, 1.51it/s] 8%|██████▍ | 11/135 [00:18<01:13, 1.69it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.1.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 8%|██████▍ | 11/135 [00:18<01:13, 1.69it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 8%|██████▍ | 11/135 [00:18<01:13, 1.69it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 8%|██████▍ | 11/135 [00:18<01:13, 1.69it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 8%|██████▍ | 11/135 [00:18<01:13, 1.69it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 8%|██████▍ | 11/135 [00:18<01:13, 1.69it/s] 10%|████████▏ | 14/135 [00:18<00:38, 3.12it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.10.input_layernorm.weight", shape: (2048,), dtype: float32 + 10%|████████▏ | 14/135 [00:18<00:38, 3.12it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 10%|████████▏ | 14/135 [00:18<00:38, 3.12it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 10%|████████▏ | 14/135 [00:18<00:38, 3.12it/s] 12%|█████████▎ | 16/135 [00:18<00:28, 4.12it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 12%|█████████▎ | 16/135 [00:18<00:28, 4.12it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 12%|█████████▎ | 16/135 [00:18<00:28, 4.12it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.10.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 12%|█████████▎ | 16/135 [00:18<00:28, 4.12it/s] 13%|██████████▌ | 18/135 [00:18<00:27, 4.23it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 13%|██████████▌ | 18/135 [00:19<00:27, 4.23it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 13%|██████████▌ | 18/135 [00:19<00:27, 4.23it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 13%|██████████▌ | 18/135 [00:19<00:27, 4.23it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 13%|██████████▌ | 18/135 [00:19<00:27, 4.23it/s] 15%|███████████▋ | 20/135 [00:19<00:21, 5.40it/s] [2024-03-19 01:28:14] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.11.input_layernorm.weight", shape: (2048,), dtype: float32 + 15%|███████████▋ | 20/135 [00:19<00:21, 5.40it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 15%|███████████▋ | 20/135 [00:19<00:21, 5.40it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 15%|███████████▋ | 20/135 [00:19<00:21, 5.40it/s] 16%|████████████▊ | 22/135 [00:19<00:17, 6.55it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 16%|████████████▊ | 22/135 [00:19<00:17, 6.55it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 16%|████████████▊ | 22/135 [00:19<00:17, 6.55it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.11.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 16%|████████████▊ | 22/135 [00:19<00:17, 6.55it/s] 18%|██████████████ | 24/135 [00:19<00:21, 5.26it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 18%|██████████████ | 24/135 [00:19<00:21, 5.26it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 18%|██████████████ | 24/135 [00:19<00:21, 5.26it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 18%|██████████████ | 24/135 [00:19<00:21, 5.26it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 18%|██████████████ | 24/135 [00:19<00:21, 5.26it/s] 19%|███████████████▏ | 26/135 [00:19<00:16, 6.56it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.12.input_layernorm.weight", shape: (2048,), dtype: float32 + 19%|███████████████▏ | 26/135 [00:19<00:16, 6.56it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 19%|███████████████▏ | 26/135 [00:20<00:16, 6.56it/s] [2024-03-19 01:28:15] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 19%|███████████████▏ | 26/135 [00:20<00:16, 6.56it/s] 21%|████████████████▍ | 28/135 [00:20<00:13, 7.75it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 21%|████████████████▍ | 28/135 [00:20<00:13, 7.75it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 21%|████████████████▍ | 28/135 [00:20<00:13, 7.75it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.12.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 21%|████████████████▍ | 28/135 [00:20<00:13, 7.75it/s] 22%|█████████████████▌ | 30/135 [00:20<00:20, 5.08it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 22%|█████████████████▌ | 30/135 [00:20<00:20, 5.08it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 22%|█████████████████▌ | 30/135 [00:20<00:20, 5.08it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 22%|█████████████████▌ | 30/135 [00:20<00:20, 5.08it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 22%|█████████████████▌ | 30/135 [00:20<00:20, 5.08it/s] 24%|██████████████████▋ | 32/135 [00:20<00:16, 6.36it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.13.input_layernorm.weight", shape: (2048,), dtype: float32 + 24%|██████████████████▋ | 32/135 [00:20<00:16, 6.36it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 24%|██████████████████▋ | 32/135 [00:21<00:16, 6.36it/s] [2024-03-19 01:28:16] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 24%|██████████████████▋ | 32/135 [00:21<00:16, 6.36it/s] 25%|███████████████████▉ | 34/135 [00:21<00:13, 7.51it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 25%|███████████████████▉ | 34/135 [00:21<00:13, 7.51it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 25%|███████████████████▉ | 34/135 [00:21<00:13, 7.51it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.13.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 25%|███████████████████▉ | 34/135 [00:21<00:13, 7.51it/s] 27%|█████████████████████ | 36/135 [00:21<00:17, 5.55it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 27%|█████████████████████ | 36/135 [00:21<00:17, 5.55it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 27%|█████████████████████ | 36/135 [00:21<00:17, 5.55it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 27%|█████████████████████ | 36/135 [00:21<00:17, 5.55it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 27%|█████████████████████ | 36/135 [00:21<00:17, 5.55it/s] 28%|██████████████████████▏ | 38/135 [00:21<00:14, 6.91it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.14.input_layernorm.weight", shape: (2048,), dtype: float32 + 28%|██████████████████████▏ | 38/135 [00:21<00:14, 6.91it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 28%|██████████████████████▏ | 38/135 [00:21<00:14, 6.91it/s] [2024-03-19 01:28:17] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 28%|██████████████████████▏ | 38/135 [00:21<00:14, 6.91it/s] 30%|███████████████████████▍ | 40/135 [00:21<00:11, 8.09it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 30%|███████████████████████▍ | 40/135 [00:22<00:11, 8.09it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 30%|███████████████████████▍ | 40/135 [00:22<00:11, 8.09it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.14.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 30%|███████████████████████▍ | 40/135 [00:22<00:11, 8.09it/s] 31%|████████████████████████▌ | 42/135 [00:22<00:15, 5.90it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 31%|████████████████████████▌ | 42/135 [00:22<00:15, 5.90it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 31%|████████████████████████▌ | 42/135 [00:22<00:15, 5.90it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 31%|████████████████████████▌ | 42/135 [00:22<00:15, 5.90it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 31%|████████████████████████▌ | 42/135 [00:22<00:15, 5.90it/s] 33%|█████████████████████████▋ | 44/135 [00:22<00:12, 7.28it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.15.input_layernorm.weight", shape: (2048,), dtype: float32 + 33%|█████████████████████████▋ | 44/135 [00:22<00:12, 7.28it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 33%|█████████████████████████▋ | 44/135 [00:22<00:12, 7.28it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 33%|█████████████████████████▋ | 44/135 [00:22<00:12, 7.28it/s] 34%|██████████████████████████▉ | 46/135 [00:22<00:10, 8.45it/s] [2024-03-19 01:28:18] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 34%|██████████████████████████▉ | 46/135 [00:23<00:10, 8.45it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 34%|██████████████████████████▉ | 46/135 [00:23<00:10, 8.45it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.15.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 34%|██████████████████████████▉ | 46/135 [00:23<00:10, 8.45it/s] 36%|████████████████████████████ | 48/135 [00:23<00:13, 6.22it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 36%|████████████████████████████ | 48/135 [00:23<00:13, 6.22it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 36%|████████████████████████████ | 48/135 [00:23<00:13, 6.22it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 36%|████████████████████████████ | 48/135 [00:23<00:13, 6.22it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 36%|████████████████████████████ | 48/135 [00:23<00:13, 6.22it/s] 37%|█████████████████████████████▎ | 50/135 [00:23<00:11, 7.61it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.16.input_layernorm.weight", shape: (2048,), dtype: float32 + 37%|█████████████████████████████▎ | 50/135 [00:23<00:11, 7.61it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 37%|█████████████████████████████▎ | 50/135 [00:23<00:11, 7.61it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 37%|█████████████████████████████▎ | 50/135 [00:23<00:11, 7.61it/s] 39%|██████████████████████████████▍ | 52/135 [00:23<00:09, 8.76it/s] [2024-03-19 01:28:19] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 39%|██████████████████████████████▍ | 52/135 [00:24<00:09, 8.76it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 39%|██████████████████████████████▍ | 52/135 [00:24<00:09, 8.76it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.16.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 39%|██████████████████████████████▍ | 52/135 [00:24<00:09, 8.76it/s] 40%|███████████████████████████████▌ | 54/135 [00:24<00:15, 5.17it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 40%|███████████████████████████████▌ | 54/135 [00:24<00:15, 5.17it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 40%|███████████████████████████████▌ | 54/135 [00:24<00:15, 5.17it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 40%|███████████████████████████████▌ | 54/135 [00:24<00:15, 5.17it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 40%|███████████████████████████████▌ | 54/135 [00:24<00:15, 5.17it/s] 41%|████████████████████████████████▊ | 56/135 [00:24<00:12, 6.48it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.17.input_layernorm.weight", shape: (2048,), dtype: float32 + 41%|████████████████████████████████▊ | 56/135 [00:24<00:12, 6.48it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 41%|████████████████████████████████▊ | 56/135 [00:24<00:12, 6.48it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 41%|████████████████████████████████▊ | 56/135 [00:24<00:12, 6.48it/s] 43%|█████████████████████████████████▉ | 58/135 [00:24<00:10, 7.66it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 43%|█████████████████████████████████▉ | 58/135 [00:25<00:10, 7.66it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 43%|█████████████████████████████████▉ | 58/135 [00:25<00:10, 7.66it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.17.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 43%|█████████████████████████████████▉ | 58/135 [00:25<00:10, 7.66it/s] 44%|███████████████████████████████████ | 60/135 [00:25<00:12, 6.06it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 44%|███████████████████████████████████ | 60/135 [00:25<00:12, 6.06it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 44%|███████████████████████████████████ | 60/135 [00:25<00:12, 6.06it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 44%|███████████████████████████████████ | 60/135 [00:25<00:12, 6.06it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 44%|███████████████████████████████████ | 60/135 [00:25<00:12, 6.06it/s] 46%|████████████████████████████████████▎ | 62/135 [00:25<00:09, 7.45it/s] [2024-03-19 01:28:20] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.18.input_layernorm.weight", shape: (2048,), dtype: float32 + 46%|████████████████████████████████████▎ | 62/135 [00:25<00:09, 7.45it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 46%|████████████████████████████████████▎ | 62/135 [00:25<00:09, 7.45it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 46%|████████████████████████████████████▎ | 62/135 [00:25<00:09, 7.45it/s] 47%|█████████████████████████████████████▍ | 64/135 [00:25<00:08, 8.60it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 47%|█████████████████████████████████████▍ | 64/135 [00:25<00:08, 8.60it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 47%|█████████████████████████████████████▍ | 64/135 [00:25<00:08, 8.60it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.18.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 47%|█████████████████████████████████████▍ | 64/135 [00:25<00:08, 8.60it/s] 49%|██████████████████████████████████████▌ | 66/135 [00:25<00:10, 6.74it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 49%|██████████████████████████████████████▌ | 66/135 [00:25<00:10, 6.74it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 49%|██████████████████████████████████████▌ | 66/135 [00:25<00:10, 6.74it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 49%|██████████████████████████████████████▌ | 66/135 [00:25<00:10, 6.74it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 49%|██████████████████████████████████████▌ | 66/135 [00:25<00:10, 6.74it/s] 50%|███████████████████████████████████████▊ | 68/135 [00:25<00:08, 8.15it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.19.input_layernorm.weight", shape: (2048,), dtype: float32 + 50%|███████████████████████████████████████▊ | 68/135 [00:25<00:08, 8.15it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 50%|███████████████████████████████████████▊ | 68/135 [00:26<00:08, 8.15it/s] [2024-03-19 01:28:21] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 50%|███████████████████████████████████████▊ | 68/135 [00:26<00:08, 8.15it/s] 52%|████████████████████████████████████████▉ | 70/135 [00:26<00:07, 9.24it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 52%|████████████████████████████████████████▉ | 70/135 [00:26<00:07, 9.24it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 52%|████████████████████████████████████████▉ | 70/135 [00:26<00:07, 9.24it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.19.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 52%|████████████████████████████████████████▉ | 70/135 [00:26<00:07, 9.24it/s] 53%|██████████████████████████████████████████▏ | 72/135 [00:26<00:10, 6.23it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 53%|██████████████████████████████████████████▏ | 72/135 [00:26<00:10, 6.23it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 53%|██████████████████████████████████████████▏ | 72/135 [00:26<00:10, 6.23it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 53%|██████████████████████████████████████████▏ | 72/135 [00:26<00:10, 6.23it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 53%|██████████████████████████████████████████▏ | 72/135 [00:26<00:10, 6.23it/s] 55%|███████████████████████████████████████████▎ | 74/135 [00:26<00:07, 7.63it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.2.input_layernorm.weight", shape: (2048,), dtype: float32 + 55%|██████████████████████████████████��████████▎ | 74/135 [00:26<00:07, 7.63it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 55%|███████████████████████████████████████████▎ | 74/135 [00:26<00:07, 7.63it/s] [2024-03-19 01:28:22] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 55%|███████████████████████████████████████████▎ | 74/135 [00:26<00:07, 7.63it/s] 56%|████████████████████████████████████████████▍ | 76/135 [00:26<00:06, 8.76it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 56%|████████████████████████████████████████████▍ | 76/135 [00:27<00:06, 8.76it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 56%|████████████████████████████████████████████▍ | 76/135 [00:27<00:06, 8.76it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.2.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 56%|████████████████████████████████████████████▍ | 76/135 [00:27<00:06, 8.76it/s] 58%|█████████████████████████████████████████████▋ | 78/135 [00:27<00:09, 5.81it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 58%|█████████████████████████████████████████████▋ | 78/135 [00:27<00:09, 5.81it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 58%|█████████████████████████████████████████████▋ | 78/135 [00:27<00:09, 5.81it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 58%|█████████████████████████████████████████████▋ | 78/135 [00:27<00:09, 5.81it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 58%|█████████████████████████████████████████████▋ | 78/135 [00:27<00:09, 5.81it/s] 59%|██████████████████████████████████████████████▊ | 80/135 [00:27<00:07, 7.17it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.20.input_layernorm.weight", shape: (2048,), dtype: float32 + 59%|██████████████████████████████████████████████▊ | 80/135 [00:27<00:07, 7.17it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 59%|██████████████████████████████████████████████▊ | 80/135 [00:27<00:07, 7.17it/s] [2024-03-19 01:28:23] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 59%|██████████████████████████████████████████████▊ | 80/135 [00:27<00:07, 7.17it/s] 61%|███████████████████████████████████████████████▉ | 82/135 [00:27<00:06, 8.33it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 61%|███████████████████████████████████████████████▉ | 82/135 [00:28<00:06, 8.33it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 61%|███████████████████████████████████████████████▉ | 82/135 [00:28<00:06, 8.33it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.20.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 61%|███████████████████████████████████████████████▉ | 82/135 [00:28<00:06, 8.33it/s] 62%|█████████████████████████████████████████████████▏ | 84/135 [00:28<00:08, 6.03it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 62%|█████████████████████████████████████████████████▏ | 84/135 [00:28<00:08, 6.03it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 62%|█████████████████████████████████████████████████▏ | 84/135 [00:28<00:08, 6.03it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 62%|█████████████████████████████████████████████████▏ | 84/135 [00:28<00:08, 6.03it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 62%|█████████████████████████████████████████████████▏ | 84/135 [00:28<00:08, 6.03it/s] 64%|██████████████████████████████████████████████████▎ | 86/135 [00:28<00:06, 7.41it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.21.input_layernorm.weight", shape: (2048,), dtype: float32 + 64%|██████████████████████████████████████████████████▎ | 86/135 [00:28<00:06, 7.41it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 64%|██████████████████████████████████████████████████▎ | 86/135 [00:28<00:06, 7.41it/s] [2024-03-19 01:28:24] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 64%|██████████████████████████████████████████████████▎ | 86/135 [00:28<00:06, 7.41it/s] 65%|███████████████████████████████████████████████████▍ | 88/135 [00:28<00:05, 8.57it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 65%|███████████████████████████████████████████████████▍ | 88/135 [00:29<00:05, 8.57it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 65%|███████████████████████████████████████████████████▍ | 88/135 [00:29<00:05, 8.57it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.21.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 65%|███████████████████████████████████████████████████▍ | 88/135 [00:29<00:05, 8.57it/s] 67%|████████████████████████████████████████████████████▋ | 90/135 [00:29<00:09, 4.99it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 67%|████████████████████████████████████████████████████▋ | 90/135 [00:29<00:09, 4.99it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 67%|████████████████████████████████████████████████████▋ | 90/135 [00:29<00:09, 4.99it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 67%|████████████████████████████████████████████████████▋ | 90/135 [00:29<00:09, 4.99it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 67%|████████████████████████████████████████████████████▋ | 90/135 [00:29<00:09, 4.99it/s] 68%|█████████████████████████████████████████████████████▊ | 92/135 [00:29<00:06, 6.29it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.3.input_layernorm.weight", shape: (2048,), dtype: float32 + 68%|█████████████████████████████████████████████████████▊ | 92/135 [00:29<00:06, 6.29it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 68%|█████████████████████████████████████████████████████▊ | 92/135 [00:29<00:06, 6.29it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 68%|█████████████████████████████████████████████████████▊ | 92/135 [00:29<00:06, 6.29it/s] 70%|███████████████████████████████████████████████████████ | 94/135 [00:29<00:05, 7.49it/s] [2024-03-19 01:28:25] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 70%|███████████████████████████████████████████████████████ | 94/135 [00:30<00:05, 7.49it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 70%|███████████████████████████████████████████████████████ | 94/135 [00:30<00:05, 7.49it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.3.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 70%|███████████████████████████████████████████████████████ | 94/135 [00:30<00:05, 7.49it/s] 71%|████████████████████████████████████████████████████████▏ | 96/135 [00:30<00:07, 5.50it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 71%|█████████████████████████████████████████████���██████████▏ | 96/135 [00:30<00:07, 5.50it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 71%|████████████████████████████████████████████████████████▏ | 96/135 [00:30<00:07, 5.50it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 71%|████████████████████████████████████████████████████████▏ | 96/135 [00:30<00:07, 5.50it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 71%|████████████████████████████████████████████████████████▏ | 96/135 [00:30<00:07, 5.50it/s] 73%|█████████████████████████████████████████████████████████▎ | 98/135 [00:30<00:05, 6.84it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.4.input_layernorm.weight", shape: (2048,), dtype: float32 + 73%|█████████████████████████████████████████████████████████▎ | 98/135 [00:30<00:05, 6.84it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 73%|█████████████████████████████████████████████████████████▎ | 98/135 [00:30<00:05, 6.84it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 73%|█████████████████████████████████████████████████████████▎ | 98/135 [00:30<00:05, 6.84it/s] 74%|█████████████████████████████████████████████████████████▊ | 100/135 [00:30<00:04, 8.04it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 74%|█████████████████████████████████████████████████████████▊ | 100/135 [00:31<00:04, 8.04it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 74%|█████████████████████████████████████████████████████████▊ | 100/135 [00:31<00:04, 8.04it/s] [2024-03-19 01:28:26] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.4.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 74%|███████████████████████████████████████████████��█████████▊ | 100/135 [00:31<00:04, 8.04it/s] 76%|██████████████████████████████████████████████████████████▉ | 102/135 [00:31<00:06, 5.44it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 76%|██████████████████████████████████████████████████████████▉ | 102/135 [00:31<00:06, 5.44it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 76%|██████████████████████████████████████████████████████████▉ | 102/135 [00:31<00:06, 5.44it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 76%|██████████████████████████████████████████████████████████▉ | 102/135 [00:31<00:06, 5.44it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 76%|██████████████████████████████████████████████████████████▉ | 102/135 [00:31<00:06, 5.44it/s] 77%|████████████████████████████████████████████████████████████ | 104/135 [00:31<00:04, 6.78it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.5.input_layernorm.weight", shape: (2048,), dtype: float32 + 77%|████████████████████████████████████████████████████████████ | 104/135 [00:31<00:04, 6.78it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 77%|████████████████████████████████████████████████████████████ | 104/135 [00:31<00:04, 6.78it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 77%|████████████████████████████████████████████████████████████ | 104/135 [00:31<00:04, 6.78it/s] 79%|█████████████████████████████████████████████████████████████▏ | 106/135 [00:31<00:03, 7.92it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 79%|█████████████████████████████████████████████████████████████▏ | 106/135 [00:31<00:03, 7.92it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 79%|█████████████████████████████████████████████████████████████▏ | 106/135 [00:32<00:03, 7.92it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.5.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 79%|█████████████████████████████████████████████████████████████▏ | 106/135 [00:32<00:03, 7.92it/s] 80%|██████████████████████████████████████████████████████████████▍ | 108/135 [00:32<00:04, 6.09it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 80%|██████████████████████████████████████████████████████████████▍ | 108/135 [00:32<00:04, 6.09it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 80%|██████████████████████████████████████████████████████████████▍ | 108/135 [00:32<00:04, 6.09it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 80%|██████████████████████████████████████████████████████████████▍ | 108/135 [00:32<00:04, 6.09it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 80%|██████████████████████████████████████████████████████████████▍ | 108/135 [00:32<00:04, 6.09it/s] 81%|███████████████████████████████████████████████████████████████▌ | 110/135 [00:32<00:03, 7.43it/s] [2024-03-19 01:28:27] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.6.input_layernorm.weight", shape: (2048,), dtype: float32 + 81%|███████████████████████████████████████████████████████████████▌ | 110/135 [00:32<00:03, 7.43it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 81%|███████████████████████████████████████████████████████████████▌ | 110/135 [00:32<00:03, 7.43it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 81%|███████████████████████████████████████████████████████████████▌ | 110/135 [00:32<00:03, 7.43it/s] 83%|████████████��███████████████████████████████████████████████████▋ | 112/135 [00:32<00:02, 8.50it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 83%|████████████████████████████████████████████████████████████████▋ | 112/135 [00:32<00:02, 8.50it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 83%|████████████████████████████████████████████████████████████████▋ | 112/135 [00:32<00:02, 8.50it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.6.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 83%|████████████████████████████████████████████████████████████████▋ | 112/135 [00:32<00:02, 8.50it/s] 84%|█████████████████████████████████████████████████████████████████▊ | 114/135 [00:32<00:02, 7.09it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 84%|█████████████████████████████████████████████████████████████████▊ | 114/135 [00:32<00:02, 7.09it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 84%|█████████████████████████████████████████████████████████████████▊ | 114/135 [00:32<00:02, 7.09it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 84%|█████████████████████████████████████████████████████████████████▊ | 114/135 [00:32<00:02, 7.09it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 84%|█████████████████████████████████████████████████████████████████▊ | 114/135 [00:32<00:02, 7.09it/s] 86%|███████████████████████████████████████████████████████████████████ | 116/135 [00:32<00:02, 8.52it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.7.input_layernorm.weight", shape: (2048,), dtype: float32 + 86%|███████████████████████████████████████████████████████████████████ | 116/135 [00:32<00:02, 8.52it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 86%|███████████████████████████████████████████████████████████████████ | 116/135 [00:32<00:02, 8.52it/s] [2024-03-19 01:28:28] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 86%|███████████████████████████████████████████████████████████████████ | 116/135 [00:32<00:02, 8.52it/s] 87%|████████████████████████████████████████████████████████████████████▏ | 118/135 [00:32<00:01, 9.59it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 87%|████████████████████████████████████████████████████████████████████▏ | 118/135 [00:33<00:01, 9.59it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 87%|████████████████████████████████████████████████████████████████████▏ | 118/135 [00:33<00:01, 9.59it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.7.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 87%|████████████████████████████████████████████████████████████████████▏ | 118/135 [00:33<00:01, 9.59it/s] 89%|█████████████████████████████████████████████████████████████████████▎ | 120/135 [00:33<00:01, 7.64it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 89%|█████████████████████████████████████████████████████████████████████▎ | 120/135 [00:33<00:01, 7.64it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 89%|█████████████████████████████████████████████████████████████████████▎ | 120/135 [00:33<00:01, 7.64it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 89%|█████████████████████████████████████████████████████████████████████▎ | 120/135 [00:33<00:01, 7.64it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 89%|█████████████████████████████████████████████████��███████████████████▎ | 120/135 [00:33<00:01, 7.64it/s] 90%|██████████████████████████████████████████████████████████████████████▍ | 122/135 [00:33<00:01, 9.07it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.8.input_layernorm.weight", shape: (2048,), dtype: float32 + 90%|██████████████████████████████████████████████████████████████████████▍ | 122/135 [00:33<00:01, 9.07it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 90%|██████████████████████████████████████████████████████████████████████▍ | 122/135 [00:33<00:01, 9.07it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 90%|██████████████████████████████████████████████████████████████████████▍ | 122/135 [00:33<00:01, 9.07it/s] 92%|███████████████████████████████████████████████████████████████████████▋ | 124/135 [00:33<00:01, 10.07it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 92%|███████████████████████████████████████████████████████████████████████▋ | 124/135 [00:34<00:01, 10.07it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 92%|███████████████████████████████████████████████████████████████████████▋ | 124/135 [00:34<00:01, 10.07it/s] [2024-03-19 01:28:29] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.8.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 92%|███████████████████████████████████████████████████████████████████████▋ | 124/135 [00:34<00:01, 10.07it/s] 93%|████████████████████████████████████████████████████████████████████████▊ | 126/135 [00:34<00:01, 6.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 93%|████████████████████████████████████████████████████████████████████████▊ | 126/135 [00:34<00:01, 6.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 93%|██████████████████████████████████████████��█████████████████████████████▊ | 126/135 [00:34<00:01, 6.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 93%|████████████████████████████████████████████████████████████████████████▊ | 126/135 [00:34<00:01, 6.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 93%|████████████████████████████████████████████████████████████████████████▊ | 126/135 [00:34<00:01, 6.36it/s] 95%|█████████████████████████████████████████████████████████████████████████▉ | 128/135 [00:34<00:00, 7.71it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.9.input_layernorm.weight", shape: (2048,), dtype: float32 + 95%|█████████████████████████████████████████████████████████████████████████▉ | 128/135 [00:34<00:00, 7.71it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_weight", shape: (2048, 704), dtype: uint32 + 95%|█████████████████████████████████████████████████████████████████████████▉ | 128/135 [00:34<00:00, 7.71it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.mlp.down_proj.q_scale", shape: (2048, 176), dtype: float32 + 95%|█████████████████████████████████████████████████████████████████████████▉ | 128/135 [00:34<00:00, 7.71it/s] 96%|███████████████████████████████████████████████████████████████████████████ | 130/135 [00:34<00:00, 8.79it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_weight", shape: (11264, 256), dtype: uint32 + 96%|███████████████████████████████████████████████████████████████████████████ | 130/135 [00:34<00:00, 8.79it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.mlp.gate_up_proj.q_scale", shape: (11264, 64), dtype: float32 + 96%|███████████████████████████████████████████████████████████████████████████ | 130/135 [00:34<00:00, 8.79it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.layers.9.post_attention_layernorm.weight", shape: (2048,), dtype: float32 + 96%|███████████████████████████████████████████████████████████████████████████ | 130/135 [00:34<00:00, 8.79it/s] 98%|████████████████████████████████████████████████████████████████████████████▎ | 132/135 [00:34<00:00, 7.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.self_attn.qkv_proj.q_weight", shape: (2560, 256), dtype: uint32 + 98%|████████████████████████████████████████████████████████████████████████████▎ | 132/135 [00:34<00:00, 7.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.self_attn.qkv_proj.q_scale", shape: (2560, 64), dtype: float32 + 98%|████████████████████████████████████████████████████████████████████████████▎ | 132/135 [00:34<00:00, 7.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_weight", shape: (2048, 256), dtype: uint32 + 98%|████████████████████████████████████████████████████████████████████████████▎ | 132/135 [00:34<00:00, 7.36it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:164: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_scale", shape: (2048, 64), dtype: float32 + 98%|████████████████████████████████████████████████████████████████████████████▎ | 132/135 [00:34<00:00, 7.36it/s] 99%|█████████████████████████████████████████████████████████████████████████████▍| 134/135 [00:34<00:00, 8.79it/s] [2024-03-19 01:28:30] INFO huggingface_loader.py:172: [Not quantized] Parameter: "model.norm.weight", shape: (2048,), dtype: float32 + 99%|█████████████████████████████████████████████████████████████████████████████▍| 134/135 [00:34<00:00, 8.79it/s] 100%|██████████████████████████████████████████████████████████████████████████████| 135/135 [00:34<00:00, 3.86it/s] +[2024-03-19 01:28:30] INFO huggingface_loader.py:194: Unloading HF weight file: /tmp/tmpo8_dluj_/repo/model.safetensors +[2024-03-19 01:28:30] INFO stats.py:76: Time usage: HF loading: 11.095 sec; Pre-quantization mapping: 5.531 sec; Quantization: 3.180 sec +[2024-03-19 01:28:30] INFO stats.py:90: RAM usage: Peak RAM: 4.098 GB. Total bytes loaded from disk: 4.098 GB +[2024-03-19 01:28:30] INFO convert_weight.py:156: Parameter size after quantization: 0.641 GB +[2024-03-19 01:28:30] INFO convert_weight.py:161: Total parameters: 1,100,056,576 +[2024-03-19 01:28:30] INFO convert_weight.py:162: Bits per parameter: 5.002 +[2024-03-19 01:28:30] INFO convert_weight.py:167: Saved to directory: /tmp/tmpx3x6wnfw + +All finished, 24 total shards committed, record saved to /tmp/tmpx3x6wnfw/ndarray-cache.json +Also saved a bf16 record to /tmp/tmpx3x6wnfw/ndarray-cache-b16.json