python -m mlc_chat gen_config /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5 --quantization q4f16_1 --conv-template phi-2 --output /tmp/tmpmpi4xlkn [2023-12-28 07:35:10] INFO auto_config.py:115: Found model configuration: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/config.json [2023-12-28 07:35:10] INFO auto_config.py:151: Found model type: phi-msft. Use `--model-type` to override. [2023-12-28 07:35:10] INFO phi_model.py:59: context_window_size not found in config.json. Falling back to n_positions (2048) [2023-12-28 07:35:10] INFO gen_config.py:129: Not found tokenizer config: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/tokenizer.model [2023-12-28 07:35:10] INFO gen_config.py:127: Found tokenizer config: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/tokenizer.json. Copying to /tmp/tmpmpi4xlkn/tokenizer.json [2023-12-28 07:35:10] INFO gen_config.py:127: Found tokenizer config: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/vocab.json. Copying to /tmp/tmpmpi4xlkn/vocab.json [2023-12-28 07:35:10] INFO gen_config.py:127: Found tokenizer config: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/merges.txt. Copying to /tmp/tmpmpi4xlkn/merges.txt [2023-12-28 07:35:10] INFO gen_config.py:127: Found tokenizer config: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/added_tokens.json. Copying to /tmp/tmpmpi4xlkn/added_tokens.json [2023-12-28 07:35:10] INFO gen_config.py:127: Found tokenizer config: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/tokenizer_config.json. Copying to /tmp/tmpmpi4xlkn/tokenizer_config.json [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting pad_token_id: 0 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting bos_token_id: 1 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting eos_token_id: 2 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting temperature: 0.7 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting repetition_penalty: 1.0 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting top_p: 0.95 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting mean_gen_len: 128 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting max_gen_len: 512 [2023-12-28 07:35:10] INFO gen_config.py:69: [System default] Setting shift_fill_factor: 0.3 [2023-12-28 07:35:10] INFO gen_config.py:157: Dumping configuration file to: /tmp/tmpmpi4xlkn/mlc-chat-config.json python -m mlc_chat convert_weight /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5 --quantization q4f16_1 --source-format auto --output /tmp/tmpmpi4xlkn [2023-12-28 07:35:10] INFO auto_config.py:115: Found model configuration: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/config.json [2023-12-28 07:35:10] INFO auto_device.py:75: Found device: cuda:0 [2023-12-28 07:35:10] INFO auto_device.py:75: Found device: cuda:1 [2023-12-28 07:35:11] INFO auto_device.py:84: Not found device: rocm:0 [2023-12-28 07:35:11] INFO auto_device.py:84: Not found device: metal:0 [2023-12-28 07:35:11] INFO auto_device.py:75: Found device: vulkan:0 [2023-12-28 07:35:11] INFO auto_device.py:75: Found device: vulkan:1 [2023-12-28 07:35:11] INFO auto_device.py:75: Found device: vulkan:2 [2023-12-28 07:35:11] INFO auto_device.py:84: Not found device: opencl:0 [2023-12-28 07:35:11] INFO auto_device.py:33: Using device: cuda:0 [2023-12-28 07:35:11] INFO auto_weight.py:70: Finding weights in: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5 [2023-12-28 07:35:11] INFO auto_weight.py:129: Found source weight format: huggingface-torch. Source configuration: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/pytorch_model.bin [2023-12-28 07:35:11] INFO auto_weight.py:149: Not found Huggingface Safetensor [2023-12-28 07:35:11] INFO auto_weight.py:106: Using source weight configuration: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/pytorch_model.bin. Use `--source` to override. [2023-12-28 07:35:11] INFO auto_weight.py:110: Using source weight format: huggingface-torch. Use `--source-format` to override. [2023-12-28 07:35:11] INFO auto_config.py:151: Found model type: phi-msft. Use `--model-type` to override. [2023-12-28 07:35:11] INFO phi_model.py:59: context_window_size not found in config.json. Falling back to n_positions (2048) [2023-12-28 07:35:14] INFO huggingface_loader.py:169: Loading HF parameters from: /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/pytorch_model.bin Weight conversion with arguments: --config /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/config.json --quantization GroupQuantize(name='q4f16_1', kind='group-quant', group_size=32, quantize_dtype='int4', storage_dtype='uint32', model_dtype='float16', num_elem_per_storage=8, num_storage_per_group=4, max_int_value=7) --model-type phi-msft --device cuda:0 --source /ssd1/cfruan/mlc-llm-repos/mlc-llm-head/dist/models/phi-1_5/pytorch_model.bin --source-format huggingface-torch --output /tmp/tmpmpi4xlkn 0%| | 0/245 [00:00