diff --git "a/logs.txt" "b/logs.txt"
new file mode 100644--- /dev/null
+++ "b/logs.txt"
@@ -0,0 +1,467 @@
+/opt/conda/envs/py310/bin/python -m mlc_llm gen_config /models/Mixtral-8x7B-Instruct-v0.1 --quantization q4f16_1 --conv-template mistral_default --output /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC
+[2024-06-06 22:07:13] INFO auto_config.py:116: Found model configuration: /models/Mixtral-8x7B-Instruct-v0.1/config.json
+[2024-06-06 22:07:13] INFO auto_config.py:154: Found model type: mixtral. Use `--model-type` to override.
+[2024-06-06 22:07:13] INFO llama_model.py:52: context_window_size not found in config.json. Falling back to max_position_embeddings (32768)
+[2024-06-06 22:07:13] INFO llama_model.py:72: prefill_chunk_size defaults to 2048
+[2024-06-06 22:07:13] INFO config.py:107: Overriding max_batch_size from 1 to 80
+[2024-06-06 22:07:13] INFO gen_config.py:143: [generation_config.json] Setting bos_token_id: 1
+[2024-06-06 22:07:13] INFO gen_config.py:143: [generation_config.json] Setting eos_token_id: 2
+[2024-06-06 22:07:13] INFO gen_config.py:155: Found tokenizer config: /models/Mixtral-8x7B-Instruct-v0.1/tokenizer.model. Copying to /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC/tokenizer.model
+[2024-06-06 22:07:13] INFO gen_config.py:155: Found tokenizer config: /models/Mixtral-8x7B-Instruct-v0.1/tokenizer.json. Copying to /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC/tokenizer.json
+[2024-06-06 22:07:13] INFO gen_config.py:157: Not found tokenizer config: /models/Mixtral-8x7B-Instruct-v0.1/vocab.json
+[2024-06-06 22:07:13] INFO gen_config.py:157: Not found tokenizer config: /models/Mixtral-8x7B-Instruct-v0.1/merges.txt
+[2024-06-06 22:07:13] INFO gen_config.py:157: Not found tokenizer config: /models/Mixtral-8x7B-Instruct-v0.1/added_tokens.json
+[2024-06-06 22:07:13] INFO gen_config.py:155: Found tokenizer config: /models/Mixtral-8x7B-Instruct-v0.1/tokenizer_config.json. Copying to /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC/tokenizer_config.json
+[2024-06-06 22:07:13] INFO gen_config.py:216: Detected tokenizer info: {'token_postproc_method': 'byte_fallback', 'prepend_space_in_encode': True, 'strip_space_in_decode': True}
+[2024-06-06 22:07:13] INFO gen_config.py:32: [System default] Setting pad_token_id: 0
+[2024-06-06 22:07:13] INFO gen_config.py:32: [System default] Setting temperature: 1.0
+[2024-06-06 22:07:13] INFO gen_config.py:32: [System default] Setting presence_penalty: 0.0
+[2024-06-06 22:07:13] INFO gen_config.py:32: [System default] Setting frequency_penalty: 0.0
+[2024-06-06 22:07:13] INFO gen_config.py:32: [System default] Setting repetition_penalty: 1.0
+[2024-06-06 22:07:13] INFO gen_config.py:32: [System default] Setting top_p: 1.0
+[2024-06-06 22:07:13] INFO gen_config.py:223: Dumping configuration file to: /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC/mlc-chat-config.json
+/opt/conda/envs/py310/bin/python -m mlc_llm convert_weight /models/Mixtral-8x7B-Instruct-v0.1 --quantization q4f16_1 --output /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC
+[2024-06-06 22:07:15] INFO auto_config.py:116: Found model configuration: /models/Mixtral-8x7B-Instruct-v0.1/config.json
+[2024-06-06 22:07:16] INFO auto_device.py:79: Found device: cuda:0
+[2024-06-06 22:07:18] INFO auto_device.py:88: Not found device: rocm:0
+[2024-06-06 22:07:19] INFO auto_device.py:88: Not found device: metal:0
+[2024-06-06 22:07:21] INFO auto_device.py:79: Found device: vulkan:0
+[2024-06-06 22:07:21] INFO auto_device.py:79: Found device: vulkan:1
+[2024-06-06 22:07:21] INFO auto_device.py:79: Found device: vulkan:2
+[2024-06-06 22:07:21] INFO auto_device.py:79: Found device: vulkan:3
+[2024-06-06 22:07:22] INFO auto_device.py:88: Not found device: opencl:0
+[2024-06-06 22:07:22] INFO auto_device.py:35: Using device: cuda:0
+[2024-06-06 22:07:22] INFO auto_weight.py:71: Finding weights in: /models/Mixtral-8x7B-Instruct-v0.1
+[2024-06-06 22:07:22] INFO auto_weight.py:137: Not found Huggingface PyTorch
+[2024-06-06 22:07:22] INFO auto_weight.py:144: Found source weight format: huggingface-safetensor. Source configuration: /models/Mixtral-8x7B-Instruct-v0.1/model.safetensors.index.json
+[2024-06-06 22:07:22] INFO auto_weight.py:107: Using source weight configuration: /models/Mixtral-8x7B-Instruct-v0.1/model.safetensors.index.json. Use `--source` to override.
+[2024-06-06 22:07:22] INFO auto_weight.py:111: Using source weight format: huggingface-safetensor. Use `--source-format` to override.
+[2024-06-06 22:07:22] INFO auto_config.py:154: Found model type: mixtral. Use `--model-type` to override.
+[2024-06-06 22:07:22] INFO llama_model.py:52: context_window_size not found in config.json. Falling back to max_position_embeddings (32768)
+[2024-06-06 22:07:22] INFO llama_model.py:72: prefill_chunk_size defaults to 2048
+Weight conversion with arguments:
+  --config          /models/Mixtral-8x7B-Instruct-v0.1/config.json
+  --quantization    GroupQuantize(name='q4f16_1', kind='group-quant', group_size=32, quantize_dtype='int4', storage_dtype='uint32', model_dtype='float16', 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      mixtral
+  --device          cuda:0
+  --source          /models/Mixtral-8x7B-Instruct-v0.1/model.safetensors.index.json
+  --source-format   huggingface-safetensor
+  --output          /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC
+Start storing to cache /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC
+
  0%|          | 0/227 [00:00<?, ?it/s]
                                       
[2024-06-06 22:08:56] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00019-of-00019.safetensors
+
  0%|          | 0/227 [00:00<?, ?it/s]
                                       
[2024-06-06 22:09:02] INFO group_quantization.py:217: Compiling quantize function for key: ((32000, 4096), float16, cuda, axis=1, output_transpose=False)
+
  0%|          | 0/227 [00:05<?, ?it/s]
                                       
[2024-06-06 22:09:03] INFO huggingface_loader.py:167: [Quantized] Parameter: "lm_head.q_weight", shape: (32000, 512), dtype: uint32
+
  0%|          | 0/227 [00:06<?, ?it/s]
                                       
[2024-06-06 22:09:03] INFO huggingface_loader.py:167: [Quantized] Parameter: "lm_head.q_scale", shape: (32000, 128), dtype: float16
+
  0%|          | 0/227 [00:07<?, ?it/s]
  0%|          | 1/227 [00:07<26:44,  7.10s/it]
                                               
[2024-06-06 22:09:03] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00018-of-00019.safetensors
+
  0%|          | 1/227 [00:07<26:44,  7.10s/it]
                                               
[2024-06-06 22:09:23] INFO group_quantization.py:217: Compiling quantize function for key: ((8, 28672, 4096), float16, cuda, axis=2, output_transpose=False)
+
  0%|          | 1/227 [00:27<26:44,  7.10s/it]
                                               
[2024-06-06 22:09:24] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
  0%|          | 1/227 [00:27<26:44,  7.10s/it]
                                               
[2024-06-06 22:09:25] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
  0%|          | 1/227 [00:29<26:44,  7.10s/it]
  1%|          | 2/227 [00:29<1:00:00, 16.00s/it]
                                                 
[2024-06-06 22:09:28] INFO group_quantization.py:217: Compiling quantize function for key: ((8, 4096, 14336), float16, cuda, axis=2, output_transpose=False)
+
  1%|          | 2/227 [00:31<1:00:00, 16.00s/it]
                                                 
[2024-06-06 22:09:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
  1%|          | 2/227 [00:31<1:00:00, 16.00s/it]
                                                 
[2024-06-06 22:09:29] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
  1%|          | 2/227 [00:32<1:00:00, 16.00s/it]
  1%|▏         | 3/227 [00:32<38:11, 10.23s/it]  
                                               
[2024-06-06 22:09:29] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.30.input_layernorm.weight", shape: (4096,), dtype: float16
+
  1%|▏         | 3/227 [00:32<38:11, 10.23s/it]
                                               
[2024-06-06 22:09:29] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.30.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
  1%|▏         | 3/227 [00:32<38:11, 10.23s/it]
                                               
[2024-06-06 22:09:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
  1%|▏         | 3/227 [00:42<38:11, 10.23s/it]
                                               
[2024-06-06 22:09:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
  1%|▏         | 3/227 [00:44<38:11, 10.23s/it]
  3%|▎         | 6/227 [00:44<22:07,  6.01s/it]
                                               
[2024-06-06 22:09:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
  3%|▎         | 6/227 [00:46<22:07,  6.01s/it]
                                               
[2024-06-06 22:09:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
  3%|▎         | 6/227 [00:46<22:07,  6.01s/it]
  3%|▎         | 7/227 [00:46<19:04,  5.20s/it]
                                               
[2024-06-06 22:09:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.31.moe.gate.weight", shape: (8, 4096), dtype: float16
+
  3%|▎         | 7/227 [00:46<19:04,  5.20s/it]
                                               
[2024-06-06 22:09:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.31.input_layernorm.weight", shape: (4096,), dtype: float16
+
  3%|▎         | 7/227 [00:46<19:04,  5.20s/it]
                                               
[2024-06-06 22:09:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.31.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
  3%|▎         | 7/227 [00:46<19:04,  5.20s/it]
                                               
[2024-06-06 22:09:43] INFO group_quantization.py:217: Compiling quantize function for key: ((6144, 4096), float16, cuda, axis=1, output_transpose=False)
+
  3%|▎         | 7/227 [00:47<19:04,  5.20s/it]
                                               
[2024-06-06 22:09:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
  3%|▎         | 7/227 [00:47<19:04,  5.20s/it]
                                               
[2024-06-06 22:09:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
  3%|▎         | 7/227 [00:47<19:04,  5.20s/it]
  5%|▍         | 11/227 [00:47<08:12,  2.28s/it]
                                                
[2024-06-06 22:09:44] INFO group_quantization.py:217: Compiling quantize function for key: ((4096, 4096), float16, cuda, axis=1, output_transpose=False)
+
  5%|▍         | 11/227 [00:47<08:12,  2.28s/it]
                                                
[2024-06-06 22:09:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
  5%|▍         | 11/227 [00:47<08:12,  2.28s/it]
                                                
[2024-06-06 22:09:44] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.31.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
  5%|▍         | 11/227 [00:48<08:12,  2.28s/it]
  5%|▌         | 12/227 [00:48<07:05,  1.98s/it]
                                                
[2024-06-06 22:09:44] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.norm.weight", shape: (4096,), dtype: float16
+
  5%|▌         | 12/227 [00:48<07:05,  1.98s/it]
                                                
[2024-06-06 22:09:44] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00018-of-00019.safetensors
+
  5%|▌         | 12/227 [00:48<07:05,  1.98s/it]
                                                
[2024-06-06 22:09:45] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00019-of-00019.safetensors
+
  5%|▌         | 12/227 [00:48<07:05,  1.98s/it]
                                                
[2024-06-06 22:09:45] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00001-of-00019.safetensors
+
  5%|▌         | 12/227 [00:49<07:05,  1.98s/it]
                                                
[2024-06-06 22:09:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.embed_tokens.q_weight", shape: (32000, 512), dtype: uint32
+
  5%|▌         | 12/227 [00:53<07:05,  1.98s/it]
                                                
[2024-06-06 22:09:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.embed_tokens.q_scale", shape: (32000, 128), dtype: float16
+
  5%|▌         | 12/227 [00:53<07:05,  1.98s/it]
  6%|▌         | 14/227 [00:54<08:14,  2.32s/it]
                                                
[2024-06-06 22:09:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
  6%|▌         | 14/227 [00:58<08:14,  2.32s/it]
                                                
[2024-06-06 22:09:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
  6%|▌         | 14/227 [00:59<08:14,  2.32s/it]
  7%|▋         | 15/227 [00:59<10:35,  3.00s/it]
                                                
[2024-06-06 22:09:58] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
  7%|▋         | 15/227 [01:01<10:35,  3.00s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
  7%|▋         | 15/227 [01:02<10:35,  3.00s/it]
  7%|▋         | 16/227 [01:02<10:17,  2.93s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.moe.gate.weight", shape: (8, 4096), dtype: float16
+
  7%|▋         | 16/227 [01:02<10:17,  2.93s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.input_layernorm.weight", shape: (4096,), dtype: float16
+
  7%|▋         | 16/227 [01:02<10:17,  2.93s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.0.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
  7%|▋         | 16/227 [01:02<10:17,  2.93s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
  7%|▋         | 16/227 [01:02<10:17,  2.93s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
  7%|▋         | 16/227 [01:02<10:17,  2.93s/it]
  9%|▉         | 20/227 [01:02<04:37,  1.34s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
  9%|▉         | 20/227 [01:02<04:37,  1.34s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.0.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
  9%|▉         | 20/227 [01:02<04:37,  1.34s/it]
                                                
[2024-06-06 22:09:59] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00002-of-00019.safetensors
+
  9%|▉         | 20/227 [01:02<04:37,  1.34s/it]
                                                
[2024-06-06 22:10:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
  9%|▉         | 20/227 [01:12<04:37,  1.34s/it]
                                                
[2024-06-06 22:10:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
  9%|▉         | 20/227 [01:13<04:37,  1.34s/it]
 10%|▉         | 22/227 [01:13<08:27,  2.48s/it]
                                                
[2024-06-06 22:10:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 10%|▉         | 22/227 [01:15<08:27,  2.48s/it]
                                                
[2024-06-06 22:10:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 10%|▉         | 22/227 [01:16<08:27,  2.48s/it]
 10%|█         | 23/227 [01:16<08:30,  2.50s/it]
                                                
[2024-06-06 22:10:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 10%|█         | 23/227 [01:16<08:30,  2.50s/it]
                                                
[2024-06-06 22:10:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 10%|█         | 23/227 [01:16<08:30,  2.50s/it]
                                                
[2024-06-06 22:10:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 10%|█         | 23/227 [01:16<08:30,  2.50s/it]
 11%|█         | 25/227 [01:16<05:48,  1.73s/it]
                                                
[2024-06-06 22:10:13] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 11%|█         | 25/227 [01:16<05:48,  1.73s/it]
                                                
[2024-06-06 22:10:13] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.1.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 11%|█         | 25/227 [01:16<05:48,  1.73s/it]
                                                
[2024-06-06 22:10:13] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.input_layernorm.weight", shape: (4096,), dtype: float16
+
 11%|█         | 25/227 [01:16<05:48,  1.73s/it]
                                                
[2024-06-06 22:10:13] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.1.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 11%|█         | 25/227 [01:16<05:48,  1.73s/it]
                                                
[2024-06-06 22:10:18] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 11%|█         | 25/227 [01:21<05:48,  1.73s/it]
                                                
[2024-06-06 22:10:19] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 11%|█         | 25/227 [01:23<05:48,  1.73s/it]
 13%|█▎        | 29/227 [01:23<05:41,  1.72s/it]
                                                
[2024-06-06 22:10:21] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 13%|█▎        | 29/227 [01:25<05:41,  1.72s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 13%|█▎        | 29/227 [01:25<05:41,  1.72s/it]
 13%|█▎        | 30/227 [01:25<06:05,  1.86s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 13%|█▎        | 30/227 [01:25<06:05,  1.86s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.input_layernorm.weight", shape: (4096,), dtype: float16
+
 13%|█▎        | 30/227 [01:25<06:05,  1.86s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.2.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 13%|█▎        | 30/227 [01:25<06:05,  1.86s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 13%|█▎        | 30/227 [01:25<06:05,  1.86s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 13%|█▎        | 30/227 [01:25<06:05,  1.86s/it]
 15%|█▍        | 34/227 [01:25<03:18,  1.03s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 15%|█▍        | 34/227 [01:26<03:18,  1.03s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.2.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 15%|█▍        | 34/227 [01:26<03:18,  1.03s/it]
                                                
[2024-06-06 22:10:22] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00001-of-00019.safetensors
+
 15%|█▍        | 34/227 [01:26<03:18,  1.03s/it]
                                                
[2024-06-06 22:10:23] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00003-of-00019.safetensors
+
 15%|█▍        | 34/227 [01:26<03:18,  1.03s/it]
                                                
[2024-06-06 22:10:35] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 15%|█▍        | 34/227 [01:38<03:18,  1.03s/it]
                                                
[2024-06-06 22:10:36] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 15%|█▍        | 34/227 [01:39<03:18,  1.03s/it]
 16%|█▌        | 36/227 [01:39<07:55,  2.49s/it]
                                                
[2024-06-06 22:10:38] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 16%|█▌        | 36/227 [01:41<07:55,  2.49s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 16%|█▌        | 36/227 [01:42<07:55,  2.49s/it]
 16%|█▋        | 37/227 [01:42<07:58,  2.52s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 16%|█▋        | 37/227 [01:42<07:58,  2.52s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 16%|█▋        | 37/227 [01:42<07:58,  2.52s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 16%|█▋        | 37/227 [01:42<07:58,  2.52s/it]
 17%|█▋        | 39/227 [01:42<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 17%|█▋        | 39/227 [01:42<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.3.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 17%|█▋        | 39/227 [01:42<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:39] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00003-of-00019.safetensors
+
 17%|█▋        | 39/227 [01:42<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:40] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00002-of-00019.safetensors
+
 17%|█▋        | 39/227 [01:43<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:40] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00007-of-00019.safetensors
+
 17%|█▋        | 39/227 [01:43<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:44] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00006-of-00019.safetensors
+
 17%|█▋        | 39/227 [01:48<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 17%|█▋        | 39/227 [01:59<05:36,  1.79s/it]
                                                
[2024-06-06 22:10:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 17%|█▋        | 39/227 [02:01<05:36,  1.79s/it]
 18%|█▊        | 41/227 [02:01<12:26,  4.01s/it]
                                                
[2024-06-06 22:10:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 18%|█▊        | 41/227 [02:03<12:26,  4.01s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 18%|█▊        | 41/227 [02:03<12:26,  4.01s/it]
 19%|█▊        | 42/227 [02:03<11:38,  3.78s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 19%|█▊        | 42/227 [02:03<11:38,  3.78s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 19%|█▊        | 42/227 [02:04<11:38,  3.78s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 19%|█▊        | 42/227 [02:04<11:38,  3.78s/it]
 19%|█▉        | 44/227 [02:04<07:44,  2.54s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 19%|█▉        | 44/227 [02:04<07:44,  2.54s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.10.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 19%|█▉        | 44/227 [02:04<07:44,  2.54s/it]
                                                
[2024-06-06 22:11:00] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00007-of-00019.safetensors
+
 19%|█▉        | 44/227 [02:04<07:44,  2.54s/it]
                                                
[2024-06-06 22:11:01] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00005-of-00019.safetensors
+
 19%|█▉        | 44/227 [02:04<07:44,  2.54s/it]
                                                
[2024-06-06 22:11:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 19%|█▉        | 44/227 [02:12<07:44,  2.54s/it]
                                                
[2024-06-06 22:11:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 19%|█▉        | 44/227 [02:12<07:44,  2.54s/it]
 20%|██        | 46/227 [02:12<09:31,  3.16s/it]
                                                
[2024-06-06 22:11:18] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 20%|██        | 46/227 [02:21<09:31,  3.16s/it]
                                                
[2024-06-06 22:11:19] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 20%|██        | 46/227 [02:22<09:31,  3.16s/it]
 21%|██        | 47/227 [02:23<13:23,  4.46s/it]
                                                
[2024-06-06 22:11:19] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.input_layernorm.weight", shape: (4096,), dtype: float16
+
 21%|██        | 47/227 [02:23<13:23,  4.46s/it]
                                                
[2024-06-06 22:11:19] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 21%|██        | 47/227 [02:23<13:23,  4.46s/it]
                                                
[2024-06-06 22:11:27] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 21%|██        | 47/227 [02:30<13:23,  4.46s/it]
                                                
[2024-06-06 22:11:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 21%|██        | 47/227 [02:31<13:23,  4.46s/it]
 22%|██▏       | 50/227 [02:31<11:11,  3.79s/it]
                                                
[2024-06-06 22:11:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 22%|██▏       | 50/227 [02:33<11:11,  3.79s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 22%|██▏       | 50/227 [02:34<11:11,  3.79s/it]
 22%|██▏       | 51/227 [02:34<10:34,  3.60s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 22%|██▏       | 51/227 [02:34<10:34,  3.60s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.input_layernorm.weight", shape: (4096,), dtype: float16
+
 22%|██▏       | 51/227 [02:34<10:34,  3.60s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.9.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 22%|██▏       | 51/227 [02:34<10:34,  3.60s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 22%|██▏       | 51/227 [02:34<10:34,  3.60s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 22%|██▏       | 51/227 [02:34<10:34,  3.60s/it]
 24%|██▍       | 55/227 [02:34<05:13,  1.82s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 24%|██▍       | 55/227 [02:34<05:13,  1.82s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.9.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 24%|██▍       | 55/227 [02:34<05:13,  1.82s/it]
                                                
[2024-06-06 22:11:31] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00005-of-00019.safetensors
+
 24%|██▍       | 55/227 [02:34<05:13,  1.82s/it]
                                                
[2024-06-06 22:11:32] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00006-of-00019.safetensors
+
 24%|██▍       | 55/227 [02:35<05:13,  1.82s/it]
                                                
[2024-06-06 22:11:32] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00007-of-00019.safetensors
+
 24%|██▍       | 55/227 [02:36<05:13,  1.82s/it]
                                                
[2024-06-06 22:11:36] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.input_layernorm.weight", shape: (4096,), dtype: float16
+
 24%|██▍       | 55/227 [02:40<05:13,  1.82s/it]
 25%|██▌       | 57/227 [02:40<05:45,  2.03s/it]
                                                
[2024-06-06 22:11:36] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.10.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 25%|██▌       | 57/227 [02:40<05:45,  2.03s/it]
                                                
[2024-06-06 22:11:36] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00008-of-00019.safetensors
+
 25%|██▌       | 57/227 [02:40<05:45,  2.03s/it]
                                                
[2024-06-06 22:11:47] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 25%|██▌       | 57/227 [02:51<05:45,  2.03s/it]
                                                
[2024-06-06 22:11:48] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 25%|██▌       | 57/227 [02:52<05:45,  2.03s/it]
 26%|██▌       | 59/227 [02:52<08:51,  3.17s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 26%|██▌       | 59/227 [02:54<08:51,  3.17s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 26%|██▌       | 59/227 [02:54<08:51,  3.17s/it]
 26%|██▋       | 60/227 [02:55<08:32,  3.07s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 26%|██▋       | 60/227 [02:55<08:32,  3.07s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 26%|██▋       | 60/227 [02:55<08:32,  3.07s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 26%|██▋       | 60/227 [02:55<08:32,  3.07s/it]
 27%|██▋       | 62/227 [02:55<05:49,  2.12s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 27%|██▋       | 62/227 [02:55<05:49,  2.12s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.11.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 27%|██▋       | 62/227 [02:55<05:49,  2.12s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.input_layernorm.weight", shape: (4096,), dtype: float16
+
 27%|██▋       | 62/227 [02:55<05:49,  2.12s/it]
                                                
[2024-06-06 22:11:51] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.11.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 27%|██▋       | 62/227 [02:55<05:49,  2.12s/it]
                                                
[2024-06-06 22:11:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 27%|██▋       | 62/227 [02:59<05:49,  2.12s/it]
                                                
[2024-06-06 22:11:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 27%|██▋       | 62/227 [03:00<05:49,  2.12s/it]
 29%|██▉       | 66/227 [03:00<04:47,  1.78s/it]
                                                
[2024-06-06 22:11:59] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 29%|██▉       | 66/227 [03:02<04:47,  1.78s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 29%|██▉       | 66/227 [03:03<04:47,  1.78s/it]
 30%|██▉       | 67/227 [03:03<05:04,  1.91s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 30%|██▉       | 67/227 [03:03<05:04,  1.91s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.input_layernorm.weight", shape: (4096,), dtype: float16
+
 30%|██▉       | 67/227 [03:03<05:04,  1.91s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.12.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 30%|██▉       | 67/227 [03:03<05:04,  1.91s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 30%|██▉       | 67/227 [03:03<05:04,  1.91s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 30%|██▉       | 67/227 [03:03<05:04,  1.91s/it]
 31%|███▏      | 71/227 [03:03<02:45,  1.06s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 31%|███▏      | 71/227 [03:03<02:45,  1.06s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.12.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 31%|███▏      | 71/227 [03:03<02:45,  1.06s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00007-of-00019.safetensors
+
 31%|███▏      | 71/227 [03:03<02:45,  1.06s/it]
                                                
[2024-06-06 22:12:00] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00009-of-00019.safetensors
+
 31%|███▏      | 71/227 [03:04<02:45,  1.06s/it]
                                                
[2024-06-06 22:12:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 31%|███▏      | 71/227 [03:14<02:45,  1.06s/it]
                                                
[2024-06-06 22:12:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 31%|███▏      | 71/227 [03:16<02:45,  1.06s/it]
 32%|███▏      | 73/227 [03:16<06:00,  2.34s/it]
                                                
[2024-06-06 22:12:14] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 32%|███▏      | 73/227 [03:18<06:00,  2.34s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 32%|███▏      | 73/227 [03:18<06:00,  2.34s/it]
 33%|███▎      | 74/227 [03:18<06:05,  2.39s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 33%|███▎      | 74/227 [03:18<06:05,  2.39s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 33%|███▎      | 74/227 [03:19<06:05,  2.39s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 33%|███▎      | 74/227 [03:19<06:05,  2.39s/it]
 33%|███▎      | 76/227 [03:19<04:16,  1.70s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 33%|███▎      | 76/227 [03:19<04:16,  1.70s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.13.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 33%|███▎      | 76/227 [03:19<04:16,  1.70s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.input_layernorm.weight", shape: (4096,), dtype: float16
+
 33%|███▎      | 76/227 [03:19<04:16,  1.70s/it]
                                                
[2024-06-06 22:12:15] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.13.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 33%|███▎      | 76/227 [03:19<04:16,  1.70s/it]
                                                
[2024-06-06 22:12:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 33%|███▎      | 76/227 [03:25<04:16,  1.70s/it]
                                                
[2024-06-06 22:12:23] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 33%|███▎      | 76/227 [03:27<04:16,  1.70s/it]
 35%|███▌      | 80/227 [03:27<04:33,  1.86s/it]
                                                
[2024-06-06 22:12:25] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 35%|███▌      | 80/227 [03:29<04:33,  1.86s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 35%|███▌      | 80/227 [03:29<04:33,  1.86s/it]
 36%|███▌      | 81/227 [03:29<04:47,  1.97s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 36%|███▌      | 81/227 [03:29<04:47,  1.97s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.input_layernorm.weight", shape: (4096,), dtype: float16
+
 36%|███▌      | 81/227 [03:29<04:47,  1.97s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.14.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 36%|███▌      | 81/227 [03:29<04:47,  1.97s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 36%|███▌      | 81/227 [03:29<04:47,  1.97s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 36%|███▌      | 81/227 [03:30<04:47,  1.97s/it]
 37%|███▋      | 85/227 [03:30<02:37,  1.11s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 37%|███▋      | 85/227 [03:30<02:37,  1.11s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.14.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 37%|███▋      | 85/227 [03:30<02:37,  1.11s/it]
                                                
[2024-06-06 22:12:26] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00008-of-00019.safetensors
+
 37%|███▋      | 85/227 [03:30<02:37,  1.11s/it]
                                                
[2024-06-06 22:12:27] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00010-of-00019.safetensors
+
 37%|███▋      | 85/227 [03:30<02:37,  1.11s/it]
 37%|███▋      | 85/227 [03:40<02:37,  1.11s/it]
                                                
[2024-06-06 22:12:37] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 37%|███▋      | 85/227 [03:40<02:37,  1.11s/it]
                                                
[2024-06-06 22:12:38] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 37%|███▋      | 85/227 [03:41<02:37,  1.11s/it]
 38%|███▊      | 87/227 [03:41<05:18,  2.27s/it]
                                                
[2024-06-06 22:12:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 38%|███▊      | 87/227 [03:43<05:18,  2.27s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 38%|███▊      | 87/227 [03:44<05:18,  2.27s/it]
 39%|███▉      | 88/227 [03:44<05:24,  2.33s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 39%|███▉      | 88/227 [03:44<05:24,  2.33s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 39%|███▉      | 88/227 [03:44<05:24,  2.33s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 39%|███▉      | 88/227 [03:44<05:24,  2.33s/it]
 40%|███▉      | 90/227 [03:44<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 40%|███▉      | 90/227 [03:44<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.15.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 40%|███▉      | 90/227 [03:44<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.input_layernorm.weight", shape: (4096,), dtype: float16
+
 40%|███▉      | 90/227 [03:44<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.15.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 40%|███▉      | 90/227 [03:44<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:41] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00009-of-00019.safetensors
+
 40%|███▉      | 90/227 [03:44<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:42] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00011-of-00019.safetensors
+
 40%|███▉      | 90/227 [03:45<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:52] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 40%|███▉      | 90/227 [03:55<03:48,  1.67s/it]
                                                
[2024-06-06 22:12:53] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 40%|███▉      | 90/227 [03:57<03:48,  1.67s/it]
 41%|████▏     | 94/227 [03:57<05:13,  2.36s/it]
                                                
[2024-06-06 22:12:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 41%|████▏     | 94/227 [03:59<05:13,  2.36s/it]
                                                
[2024-06-06 22:12:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 41%|████▏     | 94/227 [04:00<05:13,  2.36s/it]
 42%|████▏     | 95/227 [04:00<05:17,  2.40s/it]
                                                
[2024-06-06 22:12:56] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 42%|████▏     | 95/227 [04:00<05:17,  2.40s/it]
                                                
[2024-06-06 22:12:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 42%|████▏     | 95/227 [04:00<05:17,  2.40s/it]
                                                
[2024-06-06 22:12:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 42%|████▏     | 95/227 [04:00<05:17,  2.40s/it]
 43%|████▎     | 97/227 [04:00<03:45,  1.74s/it]
                                                
[2024-06-06 22:12:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 43%|████▎     | 97/227 [04:00<03:45,  1.74s/it]
                                                
[2024-06-06 22:12:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.16.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 43%|████▎     | 97/227 [04:00<03:45,  1.74s/it]
                                                
[2024-06-06 22:12:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.input_layernorm.weight", shape: (4096,), dtype: float16
+
 43%|████▎     | 97/227 [04:00<03:45,  1.74s/it]
                                                
[2024-06-06 22:12:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.16.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 43%|████▎     | 97/227 [04:00<03:45,  1.74s/it]
                                                
[2024-06-06 22:13:02] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 43%|████▎     | 97/227 [04:05<03:45,  1.74s/it]
                                                
[2024-06-06 22:13:03] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 43%|████▎     | 97/227 [04:07<03:45,  1.74s/it]
 44%|████▍     | 101/227 [04:07<03:38,  1.74s/it]
                                                 
[2024-06-06 22:13:05] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 44%|████▍     | 101/227 [04:09<03:38,  1.74s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 44%|████▍     | 101/227 [04:09<03:38,  1.74s/it]
 45%|████▍     | 102/227 [04:09<03:52,  1.86s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 45%|████▍     | 102/227 [04:09<03:52,  1.86s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.input_layernorm.weight", shape: (4096,), dtype: float16
+
 45%|████▍     | 102/227 [04:09<03:52,  1.86s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.17.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 45%|████▍     | 102/227 [04:09<03:52,  1.86s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 45%|████▍     | 102/227 [04:09<03:52,  1.86s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 45%|████▍     | 102/227 [04:09<03:52,  1.86s/it]
 47%|████▋     | 106/227 [04:09<02:07,  1.06s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 47%|████▋     | 106/227 [04:10<02:07,  1.06s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.17.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 47%|████▋     | 106/227 [04:10<02:07,  1.06s/it]
                                                 
[2024-06-06 22:13:06] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00010-of-00019.safetensors
+
 47%|████▋     | 106/227 [04:10<02:07,  1.06s/it]
                                                 
[2024-06-06 22:13:07] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00012-of-00019.safetensors
+
 47%|████▋     | 106/227 [04:10<02:07,  1.06s/it]
                                                 
[2024-06-06 22:13:16] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 47%|████▋     | 106/227 [04:19<02:07,  1.06s/it]
 47%|████▋     | 106/227 [04:20<02:07,  1.06s/it]
                                                 
[2024-06-06 22:13:17] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 47%|████▋     | 106/227 [04:20<02:07,  1.06s/it]
 48%|████▊     | 108/227 [04:21<04:12,  2.12s/it]
                                                 
[2024-06-06 22:13:19] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 48%|████▊     | 108/227 [04:22<04:12,  2.12s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 48%|████▊     | 108/227 [04:23<04:12,  2.12s/it]
 48%|████▊     | 109/227 [04:23<04:18,  2.19s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 48%|████▊     | 109/227 [04:23<04:18,  2.19s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 48%|████▊     | 109/227 [04:23<04:18,  2.19s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 48%|████▊     | 109/227 [04:23<04:18,  2.19s/it]
 49%|████▉     | 111/227 [04:23<03:02,  1.57s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 49%|████▉     | 111/227 [04:23<03:02,  1.57s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.18.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 49%|████▉     | 111/227 [04:23<03:02,  1.57s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.input_layernorm.weight", shape: (4096,), dtype: float16
+
 49%|████▉     | 111/227 [04:23<03:02,  1.57s/it]
                                                 
[2024-06-06 22:13:20] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.18.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 49%|████▉     | 111/227 [04:23<03:02,  1.57s/it]
                                                 
[2024-06-06 22:13:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 49%|████▉     | 111/227 [04:30<03:02,  1.57s/it]
                                                 
[2024-06-06 22:13:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 49%|████▉     | 111/227 [04:31<03:02,  1.57s/it]
 51%|█████     | 115/227 [04:31<03:17,  1.77s/it]
                                                 
[2024-06-06 22:13:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 51%|█████     | 115/227 [04:33<03:17,  1.77s/it]
                                                 
[2024-06-06 22:13:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 51%|█████     | 115/227 [04:34<03:17,  1.77s/it]
 51%|█████     | 116/227 [04:34<03:30,  1.89s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 51%|█████     | 116/227 [04:34<03:30,  1.89s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.input_layernorm.weight", shape: (4096,), dtype: float16
+
 51%|█████     | 116/227 [04:34<03:30,  1.89s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.19.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 51%|█████     | 116/227 [04:34<03:30,  1.89s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 51%|█████     | 116/227 [04:34<03:30,  1.89s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 51%|█████     | 116/227 [04:34<03:30,  1.89s/it]
 53%|█████▎    | 120/227 [04:34<01:54,  1.07s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 53%|█████▎    | 120/227 [04:34<01:54,  1.07s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.19.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 53%|█████▎    | 120/227 [04:34<01:54,  1.07s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00011-of-00019.safetensors
+
 53%|█████▎    | 120/227 [04:34<01:54,  1.07s/it]
                                                 
[2024-06-06 22:13:31] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00013-of-00019.safetensors
+
 53%|█████▎    | 120/227 [04:35<01:54,  1.07s/it]
                                                 
[2024-06-06 22:13:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 53%|█████▎    | 120/227 [04:45<01:54,  1.07s/it]
                                                 
[2024-06-06 22:13:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 53%|█████▎    | 120/227 [04:46<01:54,  1.07s/it]
 54%|█████▎    | 122/227 [04:47<04:04,  2.33s/it]
                                                 
[2024-06-06 22:13:45] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 54%|█████▎    | 122/227 [04:49<04:04,  2.33s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 54%|█████▎    | 122/227 [04:49<04:04,  2.33s/it]
 54%|█████▍    | 123/227 [04:49<04:06,  2.37s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 54%|█████▍    | 123/227 [04:49<04:06,  2.37s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 54%|█████▍    | 123/227 [04:49<04:06,  2.37s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 54%|█████▍    | 123/227 [04:49<04:06,  2.37s/it]
 55%|█████▌    | 125/227 [04:49<02:52,  1.69s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 55%|█████▌    | 125/227 [04:50<02:52,  1.69s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.20.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 55%|█████▌    | 125/227 [04:50<02:52,  1.69s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.input_layernorm.weight", shape: (4096,), dtype: float16
+
 55%|█████▌    | 125/227 [04:50<02:52,  1.69s/it]
                                                 
[2024-06-06 22:13:46] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.20.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 55%|█████▌    | 125/227 [04:50<02:52,  1.69s/it]
                                                 
[2024-06-06 22:13:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 55%|█████▌    | 125/227 [04:55<02:52,  1.69s/it]
                                                 
[2024-06-06 22:13:53] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 55%|█████▌    | 125/227 [04:56<02:52,  1.69s/it]
 57%|█████▋    | 129/227 [04:56<02:44,  1.68s/it]
                                                 
[2024-06-06 22:13:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 57%|█████▋    | 129/227 [04:58<02:44,  1.68s/it]
                                                 
[2024-06-06 22:13:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 57%|█████▋    | 129/227 [04:59<02:44,  1.68s/it]
 57%|█████▋    | 130/227 [04:59<02:56,  1.82s/it]
                                                 
[2024-06-06 22:13:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 57%|█████▋    | 130/227 [04:59<02:56,  1.82s/it]
                                                 
[2024-06-06 22:13:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.input_layernorm.weight", shape: (4096,), dtype: float16
+
 57%|█████▋    | 130/227 [04:59<02:56,  1.82s/it]
                                                 
[2024-06-06 22:13:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.21.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 57%|█████▋    | 130/227 [04:59<02:56,  1.82s/it]
                                                 
[2024-06-06 22:13:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 57%|█████▋    | 130/227 [04:59<02:56,  1.82s/it]
                                                 
[2024-06-06 22:13:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 57%|█████▋    | 130/227 [04:59<02:56,  1.82s/it]
 59%|█████▉    | 134/227 [04:59<01:35,  1.03s/it]
                                                 
[2024-06-06 22:13:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 59%|█████▉    | 134/227 [04:59<01:35,  1.03s/it]
                                                 
[2024-06-06 22:13:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.21.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 59%|█████▉    | 134/227 [04:59<01:35,  1.03s/it]
                                                 
[2024-06-06 22:13:56] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00012-of-00019.safetensors
+
 59%|█████▉    | 134/227 [04:59<01:35,  1.03s/it]
                                                 
[2024-06-06 22:13:56] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00014-of-00019.safetensors
+
 59%|█████▉    | 134/227 [05:00<01:35,  1.03s/it]
 59%|█████▉    | 134/227 [05:10<01:35,  1.03s/it]
                                                 
[2024-06-06 22:14:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 59%|█████▉    | 134/227 [05:11<01:35,  1.03s/it]
                                                 
[2024-06-06 22:14:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 59%|█████▉    | 134/227 [05:12<01:35,  1.03s/it]
 60%|█████▉    | 136/227 [05:12<03:34,  2.35s/it]
                                                 
[2024-06-06 22:14:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 60%|█████▉    | 136/227 [05:14<03:34,  2.35s/it]
                                                 
[2024-06-06 22:14:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 60%|█████▉    | 136/227 [05:15<03:34,  2.35s/it]
 60%|██████    | 137/227 [05:15<03:35,  2.40s/it]
                                                 
[2024-06-06 22:14:11] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 60%|██████    | 137/227 [05:15<03:35,  2.40s/it]
                                                 
[2024-06-06 22:14:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 60%|██████    | 137/227 [05:15<03:35,  2.40s/it]
                                                 
[2024-06-06 22:14:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 60%|██████    | 137/227 [05:15<03:35,  2.40s/it]
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.22.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.input_layernorm.weight", shape: (4096,), dtype: float16
+
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.22.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:12] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00013-of-00019.safetensors
+
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:12] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00015-of-00019.safetensors
+
 61%|██████    | 139/227 [05:15<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:23] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 61%|██████    | 139/227 [05:27<02:30,  1.71s/it]
                                                 
[2024-06-06 22:14:25] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 61%|██████    | 139/227 [05:28<02:30,  1.71s/it]
 63%|██████▎   | 143/227 [05:28<03:26,  2.46s/it]
                                                 
[2024-06-06 22:14:27] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 63%|██████▎   | 143/227 [05:30<03:26,  2.46s/it]
                                                 
[2024-06-06 22:14:27] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 63%|██████▎   | 143/227 [05:31<03:26,  2.46s/it]
 63%|██████▎   | 144/227 [05:31<03:26,  2.49s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 63%|██████▎   | 144/227 [05:31<03:26,  2.49s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 63%|██████▎   | 144/227 [05:31<03:26,  2.49s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 63%|██████▎   | 144/227 [05:31<03:26,  2.49s/it]
 64%|██████▍   | 146/227 [05:31<02:25,  1.80s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 64%|██���███▍   | 146/227 [05:31<02:25,  1.80s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.23.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 64%|██████▍   | 146/227 [05:31<02:25,  1.80s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.input_layernorm.weight", shape: (4096,), dtype: float16
+
 64%|██████▍   | 146/227 [05:31<02:25,  1.80s/it]
                                                 
[2024-06-06 22:14:28] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.23.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 64%|██████▍   | 146/227 [05:31<02:25,  1.80s/it]
                                                 
[2024-06-06 22:14:34] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 64%|██████▍   | 146/227 [05:38<02:25,  1.80s/it]
                                                 
[2024-06-06 22:14:36] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 64%|██████▍   | 146/227 [05:39<02:25,  1.80s/it]
 66%|██████▌   | 150/227 [05:39<02:26,  1.90s/it]
                                                 
[2024-06-06 22:14:38] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 66%|██████▌   | 150/227 [05:41<02:26,  1.90s/it]
                                                 
[2024-06-06 22:14:38] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 66%|██████▌   | 150/227 [05:42<02:26,  1.90s/it]
 67%|██████▋   | 151/227 [05:42<02:31,  2.00s/it]
                                                 
[2024-06-06 22:14:38] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.24.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 67%|██████▋   | 151/227 [05:42<02:31,  2.00s/it]
                                                 
[2024-06-06 22:14:38] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.24.input_layernorm.weight", shape: (4096,), dtype: float16
+
 67%|██████▋   | 151/227 [05:42<02:31,  2.00s/it]
                                                 
[2024-06-06 22:14:38] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.24.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 67%|██████▋   | 151/227 [05:42<02:31,  2.00s/it]
                                                 
[2024-06-06 22:14:38] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 67%|██████▋   | 151/227 [05:42<02:31,  2.00s/it]
                                                 
[2024-06-06 22:14:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 67%|██████▋   | 151/227 [05:42<02:31,  2.00s/it]
 68%|██████▊   | 155/227 [05:42<01:21,  1.13s/it]
                                                 
[2024-06-06 22:14:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 68%|██████▊   | 155/227 [05:42<01:21,  1.13s/it]
                                                 
[2024-06-06 22:14:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.24.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 68%|██████▊   | 155/227 [05:42<01:21,  1.13s/it]
                                                 
[2024-06-06 22:14:39] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00014-of-00019.safetensors
+
 68%|██████▊   | 155/227 [05:42<01:21,  1.13s/it]
                                                 
[2024-06-06 22:14:39] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00016-of-00019.safetensors
+
 68%|██████▊   | 155/227 [05:42<01:21,  1.13s/it]
                                                 
[2024-06-06 22:14:53] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 68%|██████▊   | 155/227 [05:56<01:21,  1.13s/it]
                                                 
[2024-06-06 22:14:54] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 68%|██████▊   | 155/227 [05:57<01:21,  1.13s/it]
 69%|██████▉   | 157/227 [05:57<03:11,  2.74s/it]
                                                 
[2024-06-06 22:14:56] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 69%|██████▉   | 157/227 [05:59<03:11,  2.74s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 69%|██████▉   | 157/227 [06:00<03:11,  2.74s/it]
 70%|██████▉   | 158/227 [06:00<03:08,  2.73s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.25.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 70%|██████▉   | 158/227 [06:00<03:08,  2.73s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 70%|██████▉   | 158/227 [06:00<03:08,  2.73s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 70%|██████▉   | 158/227 [06:00<03:08,  2.73s/it]
 70%|███████   | 160/227 [06:00<02:10,  1.95s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 70%|███████   | 160/227 [06:00<02:10,  1.95s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.25.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 70%|███████   | 160/227 [06:00<02:10,  1.95s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.25.input_layernorm.weight", shape: (4096,), dtype: float16
+
 70%|███████   | 160/227 [06:00<02:10,  1.95s/it]
                                                 
[2024-06-06 22:14:57] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.25.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 70%|███████   | 160/227 [06:00<02:10,  1.95s/it]
                                                 
[2024-06-06 22:15:04] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 70%|███████   | 160/227 [06:08<02:10,  1.95s/it]
                                                 
[2024-06-06 22:15:06] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 70%|███████   | 160/227 [06:09<02:10,  1.95s/it]
 72%|███████▏  | 164/227 [06:09<02:10,  2.07s/it]
                                                 
[2024-06-06 22:15:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 72%|███████▏  | 164/227 [06:11<02:10,  2.07s/it]
                                                 
[2024-06-06 22:15:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 72%|███████▏  | 164/227 [06:12<02:10,  2.07s/it]
 73%|███████▎  | 165/227 [06:12<02:13,  2.15s/it]
                                                 
[2024-06-06 22:15:08] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.26.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 73%|███████▎  | 165/227 [06:12<02:13,  2.15s/it]
                                                 
[2024-06-06 22:15:08] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.26.input_layernorm.weight", shape: (4096,), dtype: float16
+
 73%|███████▎  | 165/227 [06:12<02:13,  2.15s/it]
                                                 
[2024-06-06 22:15:08] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.26.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 73%|███████▎  | 165/227 [06:12<02:13,  2.15s/it]
                                                 
[2024-06-06 22:15:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 73%|███████▎  | 165/227 [06:12<02:13,  2.15s/it]
                                                 
[2024-06-06 22:15:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 73%|███████▎  | 165/227 [06:12<02:13,  2.15s/it]
 74%|███████▍  | 169/227 [06:12<01:10,  1.21s/it]
                                                 
[2024-06-06 22:15:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 74%|███████▍  | 169/227 [06:12<01:10,  1.21s/it]
                                                 
[2024-06-06 22:15:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.26.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 74%|███████▍  | 169/227 [06:12<01:10,  1.21s/it]
                                                 
[2024-06-06 22:15:09] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00015-of-00019.safetensors
+
 74%|███████▍  | 169/227 [06:12<01:10,  1.21s/it]
                                                 
[2024-06-06 22:15:09] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00017-of-00019.safetensors
+
 74%|███████▍  | 169/227 [06:13<01:10,  1.21s/it]
                                                 
[2024-06-06 22:15:22] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 74%|███████▍  | 169/227 [06:25<01:10,  1.21s/it]
                                                 
[2024-06-06 22:15:23] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 74%|███████▍  | 169/227 [06:26<01:10,  1.21s/it]
 75%|███████▌  | 171/227 [06:27<02:30,  2.69s/it]
                                                 
[2024-06-06 22:15:25] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 75%|███████▌  | 171/227 [06:29<02:30,  2.69s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 75%|███████▌  | 171/227 [06:29<02:30,  2.69s/it]
 76%|███████▌  | 172/227 [06:29<02:27,  2.68s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.27.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 76%|███████▌  | 172/227 [06:29<02:27,  2.68s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 76%|███████▌  | 172/227 [06:29<02:27,  2.68s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 76%|███████▌  | 172/227 [06:29<02:27,  2.68s/it]
 77%|███████▋  | 174/227 [06:29<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 77%|███████▋  | 174/227 [06:30<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.27.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 77%|███████▋  | 174/227 [06:30<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.27.input_layernorm.weight", shape: (4096,), dtype: float16
+
 77%|███████▋  | 174/227 [06:30<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.27.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 77%|███████▋  | 174/227 [06:30<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:26] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00016-of-00019.safetensors
+
 77%|███████▋  | 174/227 [06:30<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:27] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00018-of-00019.safetensors
+
 77%|███████▋  | 174/227 [06:30<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:39] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 77%|███████▋  | 174/227 [06:42<01:41,  1.91s/it]
                                                 
[2024-06-06 22:15:40] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 77%|███████▋  | 174/227 [06:43<01:41,  1.91s/it]
 78%|███████▊  | 178/227 [06:43<02:08,  2.62s/it]
                                                 
[2024-06-06 22:15:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 78%|███████▊  | 178/227 [06:45<02:08,  2.62s/it]
                                                 
[2024-06-06 22:15:42] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 78%|███████▊  | 178/227 [06:46<02:08,  2.62s/it]
 79%|███████▉  | 179/227 [06:46<02:05,  2.62s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.28.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 79%|███████▉  | 179/227 [06:46<02:05,  2.62s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 79%|███████▉  | 179/227 [06:46<02:05,  2.62s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 79%|███████▉  | 179/227 [06:46<02:05,  2.62s/it]
 80%|███████▉  | 181/227 [06:46<01:26,  1.89s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 80%|███████▉  | 181/227 [06:46<01:26,  1.89s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.28.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 80%|███████▉  | 181/227 [06:46<01:26,  1.89s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.28.input_layernorm.weight", shape: (4096,), dtype: float16
+
 80%|███████▉  | 181/227 [06:46<01:26,  1.89s/it]
                                                 
[2024-06-06 22:15:43] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.28.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 80%|███████▉  | 181/227 [06:46<01:26,  1.89s/it]
                                                 
[2024-06-06 22:15:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 80%|███████▉  | 181/227 [06:54<01:26,  1.89s/it]
                                                 
[2024-06-06 22:15:52] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 80%|███████▉  | 181/227 [06:55<01:26,  1.89s/it]
 81%|████████▏ | 185/227 [06:56<01:28,  2.11s/it]
                                                 
[2024-06-06 22:15:54] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 81%|████████▏ | 185/227 [06:57<01:28,  2.11s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 81%|████████▏ | 185/227 [06:58<01:28,  2.11s/it]
 82%|████████▏ | 186/227 [06:58<01:29,  2.19s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.29.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 82%|████████▏ | 186/227 [06:58<01:29,  2.19s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.29.input_layernorm.weight", shape: (4096,), dtype: float16
+
 82%|████████▏ | 186/227 [06:58<01:29,  2.19s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.29.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 82%|████████▏ | 186/227 [06:58<01:29,  2.19s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 82%|████████▏ | 186/227 [06:58<01:29,  2.19s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 82%|████████▏ | 186/227 [06:58<01:29,  2.19s/it]
 84%|████████▎ | 190/227 [06:58<00:45,  1.24s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 84%|████████▎ | 190/227 [06:58<00:45,  1.24s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.29.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 84%|████████▎ | 190/227 [06:58<00:45,  1.24s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.30.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 84%|████████▎ | 190/227 [06:58<00:45,  1.24s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 84%|████████▎ | 190/227 [06:59<00:45,  1.24s/it]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 84%|████████▎ | 190/227 [06:59<00:45,  1.24s/it]
 85%|████████▌ | 193/227 [06:59<00:29,  1.16it/s]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 85%|████████▌ | 193/227 [06:59<00:29,  1.16it/s]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.30.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 85%|████████▌ | 193/227 [06:59<00:29,  1.16it/s]
                                                 
[2024-06-06 22:15:55] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00017-of-00019.safetensors
+
 85%|████████▌ | 193/227 [06:59<00:29,  1.16it/s]
                                                 
[2024-06-06 22:15:56] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00018-of-00019.safetensors
+
 85%|████████▌ | 193/227 [06:59<00:29,  1.16it/s]
                                                 
[2024-06-06 22:15:56] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00003-of-00019.safetensors
+
 85%|████████▌ | 193/227 [07:00<00:29,  1.16it/s]
                                                 
[2024-06-06 22:16:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.input_layernorm.weight", shape: (4096,), dtype: float16
+
 85%|████████▌ | 193/227 [07:04<00:29,  1.16it/s]
 86%|████████▌ | 195/227 [07:04<00:42,  1.33s/it]
                                                 
[2024-06-06 22:16:01] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.3.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 86%|████████▌ | 195/227 [07:04<00:42,  1.33s/it]
                                                 
[2024-06-06 22:16:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 86%|████████▌ | 195/227 [07:10<00:42,  1.33s/it]
                                                 
[2024-06-06 22:16:08] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 86%|█████���██▌ | 195/227 [07:11<00:42,  1.33s/it]
 87%|████████▋ | 197/227 [07:11<00:56,  1.89s/it]
                                                 
[2024-06-06 22:16:10] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 87%|████████▋ | 197/227 [07:13<00:56,  1.89s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 87%|████████▋ | 197/227 [07:14<00:56,  1.89s/it]
 87%|████████▋ | 198/227 [07:14<00:58,  2.01s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 87%|████████▋ | 198/227 [07:14<00:58,  2.01s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.input_layernorm.weight", shape: (4096,), dtype: float16
+
 87%|████████▋ | 198/227 [07:14<00:58,  2.01s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.4.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 87%|████████▋ | 198/227 [07:14<00:58,  2.01s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 87%|████████▋ | 198/227 [07:14<00:58,  2.01s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 87%|████████▋ | 198/227 [07:14<00:58,  2.01s/it]
 89%|████████▉ | 202/227 [07:14<00:27,  1.09s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 89%|████████▉ | 202/227 [07:14<00:27,  1.09s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.4.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 89%|████████▉ | 202/227 [07:14<00:27,  1.09s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 89%|████████▉ | 202/227 [07:14<00:27,  1.09s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 89%|████████▉ | 202/227 [07:14<00:27,  1.09s/it]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 89%|████████▉ | 202/227 [07:14<00:27,  1.09s/it]
 90%|█████████ | 205/227 [07:14<00:16,  1.33it/s]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 90%|█████████ | 205/227 [07:15<00:16,  1.33it/s]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 90%|█████████ | 205/227 [07:15<00:16,  1.33it/s]
                                                 
[2024-06-06 22:16:11] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00003-of-00019.safetensors
+
 90%|█████████ | 205/227 [07:15<00:16,  1.33it/s]
                                                 
[2024-06-06 22:16:12] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00004-of-00019.safetensors
+
 90%|█████████ | 205/227 [07:15<00:16,  1.33it/s]
 90%|█████████ | 205/227 [07:30<00:16,  1.33it/s]
                                                 
[2024-06-06 22:16:27] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 90%|█████████ | 205/227 [07:31<00:16,  1.33it/s]
                                                 
[2024-06-06 22:16:28] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 90%|█████████ | 205/227 [07:32<00:16,  1.33it/s]
 91%|█████████ | 207/227 [07:32<00:53,  2.67s/it]
                                                 
[2024-06-06 22:16:30] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 91%|█████████ | 207/227 [07:34<00:53,  2.67s/it]
                                                 
[2024-06-06 22:16:31] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.5.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 91%|█████████ | 207/227 [07:34<00:53,  2.67s/it]
 92%|█████████▏| 208/227 [07:35<00:50,  2.66s/it]
                                                 
[2024-06-06 22:16:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.input_layernorm.weight", shape: (4096,), dtype: float16
+
 92%|█████████▏| 208/227 [07:35<00:50,  2.66s/it]
                                                 
[2024-06-06 22:16:31] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.5.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 92%|█████████▏| 208/227 [07:35<00:50,  2.66s/it]
                                                 
[2024-06-06 22:16:31] INFO huggingface_loader.py:185: Loading HF parameters from: /models/Mixtral-8x7B-Instruct-v0.1/model-00005-of-00019.safetensors
+
 92%|█████████▏| 208/227 [07:35<00:50,  2.66s/it]
                                                 
[2024-06-06 22:16:50] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 92%|█████████▏| 208/227 [07:53<00:50,  2.66s/it]
                                                 
[2024-06-06 22:16:51] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 92%|█████████▏| 208/227 [07:55<00:50,  2.66s/it]
 93%|█████████▎| 211/227 [07:55<01:08,  4.26s/it]
                                                 
[2024-06-06 22:16:54] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 93%|█████████▎| 211/227 [07:58<01:08,  4.26s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 93%|█████████▎| 211/227 [07:58<01:08,  4.26s/it]
 93%|█████████▎| 212/227 [07:58<01:01,  4.12s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 93%|█████████▎| 212/227 [07:58<01:01,  4.12s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 93%|█████████▎| 212/227 [07:58<01:01,  4.12s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 93%|█████████▎| 212/227 [07:58<01:01,  4.12s/it]
 94%|█████████▍| 214/227 [07:58<00:37,  2.88s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 94%|█████████▍| 214/227 [07:59<00:37,  2.88s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.6.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 94%|█████████▍| 214/227 [07:59<00:37,  2.88s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.input_layernorm.weight", shape: (4096,), dtype: float16
+
 94%|█████████▍| 214/227 [07:59<00:37,  2.88s/it]
                                                 
[2024-06-06 22:16:55] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.6.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 94%|█████████▍| 214/227 [07:59<00:37,  2.88s/it]
                                                 
[2024-06-06 22:17:07] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.moe.e1_e3.q_weight", shape: (8, 28672, 512), dtype: uint32
+
 94%|█████████▍| 214/227 [08:11<00:37,  2.88s/it]
                                                 
[2024-06-06 22:17:09] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.moe.e1_e3.q_scale", shape: (8, 28672, 128), dtype: float16
+
 94%|█████████▍| 214/227 [08:12<00:37,  2.88s/it]
 96%|█████████▌| 218/227 [08:12<00:28,  3.14s/it]
                                                 
[2024-06-06 22:17:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.moe.e2.q_weight", shape: (8, 4096, 1792), dtype: uint32
+
 96%|█████████▌| 218/227 [08:14<00:28,  3.14s/it]
                                                 
[2024-06-06 22:17:11] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.moe.e2.q_scale", shape: (8, 4096, 448), dtype: float16
+
 96%|█████████▌| 218/227 [08:15<00:28,  3.14s/it]
 96%|█████████▋| 219/227 [08:15<00:24,  3.07s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 96%|█████████▋| 219/227 [08:15<00:24,  3.07s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.input_layernorm.weight", shape: (4096,), dtype: float16
+
 96%|█████████▋| 219/227 [08:15<00:24,  3.07s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.7.post_attention_layernorm.weight", shape: (4096,), dtype: float16
+
 96%|█████████▋| 219/227 [08:15<00:24,  3.07s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 96%|█████████▋| 219/227 [08:15<00:24,  3.07s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 96%|█████████▋| 219/227 [08:15<00:24,  3.07s/it]
 98%|█████████▊| 223/227 [08:15<00:06,  1.71s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
 98%|█████████▊| 223/227 [08:15<00:06,  1.71s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.7.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
 98%|█████████▊| 223/227 [08:15<00:06,  1.71s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:175: [Not quantized] Parameter: "model.layers.8.moe.gate.weight", shape: (8, 4096), dtype: float16
+
 98%|█████████▊| 223/227 [08:15<00:06,  1.71s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.qkv_proj.q_weight", shape: (6144, 512), dtype: uint32
+
 98%|█████████▊| 223/227 [08:15<00:06,  1.71s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.qkv_proj.q_scale", shape: (6144, 128), dtype: float16
+
 98%|█████████▊| 223/227 [08:15<00:06,  1.71s/it]
100%|█████████▉| 226/227 [08:15<00:01,  1.18s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_weight", shape: (4096, 512), dtype: uint32
+
100%|█████████▉| 226/227 [08:15<00:01,  1.18s/it]
                                                 
[2024-06-06 22:17:12] INFO huggingface_loader.py:167: [Quantized] Parameter: "model.layers.8.self_attn.o_proj.q_scale", shape: (4096, 128), dtype: float16
+
100%|█████████▉| 226/227 [08:15<00:01,  1.18s/it]
100%|██████████| 227/227 [08:15<00:00,  2.18s/it]
+[2024-06-06 22:17:12] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00004-of-00019.safetensors
+[2024-06-06 22:17:13] INFO huggingface_loader.py:197: Unloading HF weight file: /models/Mixtral-8x7B-Instruct-v0.1/model-00005-of-00019.safetensors
+[2024-06-06 22:17:13] INFO stats.py:77: Time usage: HF loading: 125.315 sec; Pre-quantization mapping: 286.394 sec; Quantization: 4.263 sec
+[2024-06-06 22:17:13] INFO stats.py:91: RAM usage: Peak RAM: 18.563 GB. Total bytes loaded from disk: 210.951 GB
+[2024-06-06 22:17:13] INFO convert_weight.py:155: Parameter size after quantization: 24.468 GB
+[2024-06-06 22:17:13] INFO convert_weight.py:160: Total parameters: 56,626,802,688
+[2024-06-06 22:17:13] INFO convert_weight.py:161: Bits per parameter: 3.712
+[2024-06-06 22:17:13] INFO convert_weight.py:166: Saved to directory: /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC
+
+All finished, 162 total shards committed, record saved to /models/mlc-delivery/hf/mlc-ai/Mixtral-8x7B-Instruct-v0.1-q4f16_1-MLC/ndarray-cache.json