Image-Text-to-Text
Transformers
Safetensors
qwen3_5
quantized
gptq
int4
qwen3.6
mtp
speculative-decoding
conversational
4-bit precision
Instructions to use palmfuture/Qwen3.6-27B-GPTQ-Int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use palmfuture/Qwen3.6-27B-GPTQ-Int4 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="palmfuture/Qwen3.6-27B-GPTQ-Int4") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# Load model directly from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("palmfuture/Qwen3.6-27B-GPTQ-Int4") model = AutoModelForImageTextToText.from_pretrained("palmfuture/Qwen3.6-27B-GPTQ-Int4") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use palmfuture/Qwen3.6-27B-GPTQ-Int4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "palmfuture/Qwen3.6-27B-GPTQ-Int4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "palmfuture/Qwen3.6-27B-GPTQ-Int4", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/palmfuture/Qwen3.6-27B-GPTQ-Int4
- SGLang
How to use palmfuture/Qwen3.6-27B-GPTQ-Int4 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "palmfuture/Qwen3.6-27B-GPTQ-Int4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "palmfuture/Qwen3.6-27B-GPTQ-Int4", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "palmfuture/Qwen3.6-27B-GPTQ-Int4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "palmfuture/Qwen3.6-27B-GPTQ-Int4", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use palmfuture/Qwen3.6-27B-GPTQ-Int4 with Docker Model Runner:
docker model run hf.co/palmfuture/Qwen3.6-27B-GPTQ-Int4
| layer,module,loss,samples,damp,time | |
| 0,linear_attn.in_proj_qkv,0.0003197445,0.05000,2.228 | |
| 0,linear_attn.in_proj_z,0.0002064770,0.05000,2.027 | |
| 0,linear_attn.out_proj,0.0000001519,0.05000,2.319 | |
| 0,mlp.up_proj,0.0000019669,0.05000,2.673 | |
| 0,mlp.gate_proj,0.0000021337,0.05000,2.799 | |
| 0,mlp.down_proj,0.0000000202,0.05000,9.369 | |
| 1,linear_attn.in_proj_qkv,0.0000158468,0.05000,1.846 | |
| 1,linear_attn.in_proj_z,0.0000097547,0.05000,1.875 | |
| 1,linear_attn.out_proj,0.0000000610,0.05000,2.394 | |
| 1,mlp.up_proj,0.0000043260,0.05000,2.832 | |
| 1,mlp.gate_proj,0.0000045451,0.05000,2.806 | |
| 1,mlp.down_proj,0.0000000292,0.05000,9.841 | |
| 2,linear_attn.in_proj_qkv,0.0000236378,0.05000,1.961 | |
| 2,linear_attn.in_proj_z,0.0000143486,0.05000,1.805 | |
| 2,linear_attn.out_proj,0.0000000818,0.05000,2.354 | |
| 2,mlp.gate_proj,0.0000083266,0.05000,2.744 | |
| 2,mlp.up_proj,0.0000078148,0.05000,2.820 | |
| 2,mlp.down_proj,0.0000001103,0.05000,9.239 | |
| 3,self_attn.k_proj,0.0000168175,0.05000,3.962 | |
| 3,self_attn.v_proj,0.0000166043,0.05000,4.016 | |
| 3,self_attn.q_proj,0.0002342821,0.05000,4.118 | |
| 3,self_attn.o_proj,0.0000000840,0.05000,2.620 | |
| 3,mlp.gate_proj,0.0000101265,0.05000,2.768 | |
| 3,mlp.up_proj,0.0000097521,0.05000,2.807 | |
| 3,mlp.down_proj,0.0000000773,0.05000,9.705 | |
| 4,linear_attn.in_proj_qkv,0.0000327987,0.05000,2.108 | |
| 4,linear_attn.in_proj_z,0.0000209418,0.05000,1.814 | |
| 4,linear_attn.out_proj,0.0000001242,0.05000,2.427 | |
| 4,mlp.gate_proj,0.0000135747,0.05000,2.789 | |
| 4,mlp.up_proj,0.0000131227,0.05000,2.835 | |
| 4,mlp.down_proj,0.0000001075,0.05000,9.531 | |
| 5,linear_attn.in_proj_qkv,0.0000415633,0.05000,1.970 | |
| 5,linear_attn.in_proj_z,0.0000271475,0.05000,1.920 | |
| 5,linear_attn.out_proj,0.0000001655,0.05000,2.505 | |
| 5,mlp.gate_proj,0.0000173205,0.05000,2.811 | |
| 5,mlp.up_proj,0.0000162889,0.05000,2.830 | |
| 5,mlp.down_proj,0.0000001692,0.05000,9.587 | |
| 6,linear_attn.in_proj_qkv,0.0000621110,0.05000,2.009 | |
| 6,linear_attn.in_proj_z,0.0000372702,0.05000,1.874 | |
| 6,linear_attn.out_proj,0.0000002171,0.05000,2.481 | |
| 6,mlp.up_proj,0.0000213310,0.05000,2.764 | |
| 6,mlp.gate_proj,0.0000229309,0.05000,3.045 | |
| 6,mlp.down_proj,0.0000002675,0.05000,9.538 | |
| 7,self_attn.v_proj,0.0000145466,0.05000,3.755 | |
| 7,self_attn.k_proj,0.0000158137,0.05000,3.900 | |
| 7,self_attn.q_proj,0.0002015561,0.05000,4.001 | |
| 7,self_attn.o_proj,0.0000001460,0.05000,2.417 | |
| 7,mlp.gate_proj,0.0000258102,0.05000,2.853 | |
| 7,mlp.up_proj,0.0000240012,0.05000,3.144 | |
| 7,mlp.down_proj,0.0000003147,0.05000,9.532 | |
| 8,linear_attn.in_proj_qkv,0.0000642539,0.05000,1.999 | |
| 8,linear_attn.in_proj_z,0.0000405329,0.05000,1.941 | |
| 8,linear_attn.out_proj,0.0000003066,0.05000,2.422 | |
| 8,mlp.up_proj,0.0000270153,0.05000,2.813 | |
| 8,mlp.gate_proj,0.0000288301,0.05000,2.838 | |
| 8,mlp.down_proj,0.0000003754,0.05000,9.611 | |
| 9,linear_attn.in_proj_qkv,0.0000657042,0.05000,1.955 | |
| 9,linear_attn.in_proj_z,0.0000400837,0.05000,1.924 | |
| 9,linear_attn.out_proj,0.0000003402,0.05000,2.335 | |
| 9,mlp.up_proj,0.0000296260,0.05000,2.857 | |
| 9,mlp.gate_proj,0.0000313970,0.05000,2.860 | |
| 9,mlp.down_proj,0.0000004274,0.05000,9.500 | |
| 10,linear_attn.in_proj_qkv,0.0000685984,0.05000,2.139 | |
| 10,linear_attn.in_proj_z,0.0000406228,0.05000,2.094 | |
| 10,linear_attn.out_proj,0.0000003843,0.05000,2.428 | |
| 10,mlp.gate_proj,0.0000321583,0.05000,2.752 | |
| 10,mlp.up_proj,0.0000305340,0.05000,2.788 | |
| 10,mlp.down_proj,0.0000004743,0.05000,9.580 | |
| 11,self_attn.k_proj,0.0000151571,0.05000,3.848 | |
| 11,self_attn.v_proj,0.0000136842,0.05000,3.913 | |
| 11,self_attn.q_proj,0.0001796321,0.05000,3.993 | |
| 11,self_attn.o_proj,0.0000002628,0.05000,2.350 | |
| 11,mlp.gate_proj,0.0000343429,0.05000,2.815 | |
| 11,mlp.up_proj,0.0000327367,0.05000,2.870 | |
| 11,mlp.down_proj,0.0000005578,0.05000,9.701 | |
| 12,linear_attn.in_proj_qkv,0.0000775487,0.05000,1.976 | |
| 12,linear_attn.in_proj_z,0.0000459151,0.05000,1.951 | |
| 12,linear_attn.out_proj,0.0000005732,0.05000,2.360 | |
| 12,mlp.gate_proj,0.0000356221,0.05000,2.867 | |
| 12,mlp.up_proj,0.0000339428,0.05000,2.894 | |
| 12,mlp.down_proj,0.0000006371,0.05000,9.814 | |
| 13,linear_attn.in_proj_qkv,0.0000816816,0.05000,2.078 | |
| 13,linear_attn.in_proj_z,0.0000461836,0.05000,2.012 | |
| 13,linear_attn.out_proj,0.0000006304,0.05000,2.461 | |
| 13,mlp.up_proj,0.0000364391,0.05000,2.754 | |
| 13,mlp.gate_proj,0.0000382620,0.05000,3.088 | |
| 13,mlp.down_proj,0.0000007334,0.05000,9.365 | |
| 14,linear_attn.in_proj_qkv,0.0000896298,0.05000,1.918 | |
| 14,linear_attn.in_proj_z,0.0000520888,0.05000,1.939 | |
| 14,linear_attn.out_proj,0.0000007492,0.05000,2.463 | |
| 14,mlp.up_proj,0.0000385812,0.05000,2.837 | |
| 14,mlp.gate_proj,0.0000403974,0.05000,2.985 | |
| 14,mlp.down_proj,0.0000008201,0.05000,9.448 | |
| 15,self_attn.v_proj,0.0000131611,0.05000,3.857 | |
| 15,self_attn.k_proj,0.0000135512,0.05000,3.920 | |
| 15,self_attn.q_proj,0.0001546285,0.05000,4.023 | |
| 15,self_attn.o_proj,0.0000004657,0.05000,2.425 | |
| 15,mlp.gate_proj,0.0000461728,0.05000,2.800 | |
| 15,mlp.up_proj,0.0000445283,0.05000,2.873 | |
| 15,mlp.down_proj,0.0000009885,0.05000,9.419 | |
| 16,linear_attn.in_proj_qkv,0.0000963353,0.05000,2.028 | |
| 16,linear_attn.in_proj_z,0.0000548140,0.05000,2.275 | |
| 16,linear_attn.out_proj,0.0000008343,0.05000,2.233 | |
| 16,mlp.gate_proj,0.0000508629,0.05000,2.775 | |
| 16,mlp.up_proj,0.0000487897,0.05000,2.822 | |
| 16,mlp.down_proj,0.0000011787,0.05000,9.420 | |
| 17,linear_attn.in_proj_qkv,0.0001095097,0.05000,2.049 | |
| 17,linear_attn.in_proj_z,0.0000560735,0.05000,2.012 | |
| 17,linear_attn.out_proj,0.0000011213,0.05000,2.254 | |
| 17,mlp.gate_proj,0.0000583781,0.05000,2.725 | |
| 17,mlp.up_proj,0.0000559772,0.05000,2.836 | |
| 17,mlp.down_proj,0.0000015447,0.05000,9.683 | |
| 18,linear_attn.in_proj_qkv,0.0001184729,0.05000,2.019 | |
| 18,linear_attn.in_proj_z,0.0000651775,0.05000,1.983 | |
| 18,linear_attn.out_proj,0.0000013323,0.05000,2.491 | |
| 18,mlp.gate_proj,0.0000716857,0.05000,2.830 | |
| 18,mlp.up_proj,0.0000672247,0.05000,2.852 | |
| 18,mlp.down_proj,0.0000024226,0.05000,9.683 | |
| 19,self_attn.k_proj,0.0000173523,0.05000,3.986 | |
| 19,self_attn.v_proj,0.0000194091,0.05000,4.015 | |
| 19,self_attn.q_proj,0.0001928319,0.05000,4.043 | |
| 19,self_attn.o_proj,0.0000018063,0.05000,2.361 | |
| 19,mlp.gate_proj,0.0000900986,0.05000,2.658 | |
| 19,mlp.up_proj,0.0000849859,0.05000,2.772 | |
| 19,mlp.down_proj,0.0000033498,0.05000,9.764 | |
| 20,linear_attn.in_proj_qkv,0.0002139989,0.05000,2.011 | |
| 20,linear_attn.in_proj_z,0.0001198187,0.05000,1.909 | |
| 20,linear_attn.out_proj,0.0000021942,0.05000,2.544 | |
| 20,mlp.gate_proj,0.0001100951,0.05000,2.911 | |
| 20,mlp.up_proj,0.0001016944,0.05000,2.926 | |
| 20,mlp.down_proj,0.0000038367,0.05000,9.316 | |
| 21,linear_attn.in_proj_qkv,0.0002620370,0.05000,2.017 | |
| 21,linear_attn.in_proj_z,0.0001673315,0.05000,1.818 | |
| 21,linear_attn.out_proj,0.0000027516,0.05000,2.425 | |
| 21,mlp.up_proj,0.0001115360,0.05000,2.836 | |
| 21,mlp.gate_proj,0.0001198745,0.05000,2.846 | |
| 21,mlp.down_proj,0.0000041294,0.05000,9.817 | |
| 22,linear_attn.in_proj_qkv,0.0002282727,0.05000,2.019 | |
| 22,linear_attn.in_proj_z,0.0001399470,0.05000,1.863 | |
| 22,linear_attn.out_proj,0.0000025874,0.05000,2.483 | |
| 22,mlp.up_proj,0.0001239583,0.05000,2.856 | |
| 22,mlp.gate_proj,0.0001402319,0.05000,2.877 | |
| 22,mlp.down_proj,0.0000046557,0.05000,9.638 | |
| 23,self_attn.k_proj,0.0000247198,0.05000,3.962 | |
| 23,self_attn.v_proj,0.0000262281,0.05000,3.982 | |
| 23,self_attn.q_proj,0.0002511130,0.05000,4.052 | |
| 23,self_attn.o_proj,0.0000022185,0.05000,2.408 | |
| 23,mlp.up_proj,0.0001334481,0.05000,2.686 | |
| 23,mlp.gate_proj,0.0001454333,0.05000,2.801 | |
| 23,mlp.down_proj,0.0000048433,0.05000,9.444 | |
| 24,linear_attn.in_proj_qkv,0.0002210956,0.05000,1.948 | |
| 24,linear_attn.in_proj_z,0.0001247617,0.05000,1.858 | |
| 24,linear_attn.out_proj,0.0000031194,0.05000,2.483 | |
| 24,mlp.up_proj,0.0001440041,0.05000,2.787 | |
| 24,mlp.gate_proj,0.0001577180,0.05000,2.968 | |
| 24,mlp.down_proj,0.0000056242,0.05000,9.495 | |
| 25,linear_attn.in_proj_qkv,0.0002246287,0.05000,2.023 | |
| 25,linear_attn.in_proj_z,0.0001186763,0.05000,1.911 | |
| 25,linear_attn.out_proj,0.0000031973,0.05000,2.360 | |
| 25,mlp.up_proj,0.0001451803,0.05000,2.870 | |
| 25,mlp.gate_proj,0.0001504104,0.05000,2.889 | |
| 25,mlp.down_proj,0.0000056412,0.05000,9.788 | |
| 26,linear_attn.in_proj_qkv,0.0002032605,0.05000,2.094 | |
| 26,linear_attn.in_proj_z,0.0001032550,0.05000,1.979 | |
| 26,linear_attn.out_proj,0.0000033678,0.05000,2.544 | |
| 26,mlp.gate_proj,0.0001398861,0.05000,2.824 | |
| 26,mlp.up_proj,0.0001431139,0.05000,2.828 | |
| 26,mlp.down_proj,0.0000059178,0.05000,9.702 | |
| 27,self_attn.v_proj,0.0000209463,0.05000,3.962 | |
| 27,self_attn.k_proj,0.0000228403,0.05000,4.013 | |
| 27,self_attn.q_proj,0.0002288074,0.05000,4.068 | |
| 27,self_attn.o_proj,0.0000044184,0.05000,2.295 | |
| 27,mlp.up_proj,0.0001511458,0.05000,2.871 | |
| 27,mlp.gate_proj,0.0001433983,0.05000,2.919 | |
| 27,mlp.down_proj,0.0000063376,0.05000,9.524 | |
| 28,linear_attn.in_proj_qkv,0.0001990004,0.05000,2.131 | |
| 28,linear_attn.in_proj_z,0.0000942670,0.05000,1.998 | |
| 28,linear_attn.out_proj,0.0000041844,0.05000,2.418 | |
| 28,mlp.up_proj,0.0001480044,0.05000,2.869 | |
| 28,mlp.gate_proj,0.0001377359,0.05000,2.896 | |
| 28,mlp.down_proj,0.0000064281,0.05000,9.705 | |
| 29,linear_attn.in_proj_qkv,0.0002153540,0.05000,1.966 | |
| 29,linear_attn.in_proj_z,0.0001002709,0.05000,2.028 | |
| 29,linear_attn.out_proj,0.0000037770,0.05000,2.416 | |
| 29,mlp.up_proj,0.0001485866,0.05000,2.798 | |
| 29,mlp.gate_proj,0.0001374452,0.05000,3.063 | |
| 29,mlp.down_proj,0.0000064556,0.05000,9.480 | |
| 30,linear_attn.in_proj_qkv,0.0002151987,0.05000,1.994 | |
| 30,linear_attn.in_proj_z,0.0001021870,0.05000,1.998 | |
| 30,linear_attn.out_proj,0.0000047097,0.05000,2.726 | |
| 30,mlp.gate_proj,0.0001347883,0.05000,2.856 | |
| 30,mlp.up_proj,0.0001493319,0.05000,2.911 | |
| 30,mlp.down_proj,0.0000066001,0.05000,9.961 | |
| 31,self_attn.k_proj,0.0000226690,0.05000,3.858 | |
| 31,self_attn.v_proj,0.0000261192,0.05000,3.924 | |
| 31,self_attn.q_proj,0.0002197505,0.05000,4.045 | |
| 31,self_attn.o_proj,0.0000056364,0.05000,2.382 | |
| 31,mlp.up_proj,0.0001621095,0.05000,2.591 | |
| 31,mlp.gate_proj,0.0001435794,0.05000,2.934 | |
| 31,mlp.down_proj,0.0000072634,0.05000,9.668 | |
| 32,linear_attn.in_proj_qkv,0.0002293290,0.05000,1.991 | |
| 32,linear_attn.in_proj_z,0.0001018995,0.05000,1.925 | |
| 32,linear_attn.out_proj,0.0000049107,0.05000,2.322 | |
| 32,mlp.up_proj,0.0001718162,0.05000,2.772 | |
| 32,mlp.gate_proj,0.0001514548,0.05000,2.775 | |
| 32,mlp.down_proj,0.0000077587,0.05000,9.593 | |
| 33,linear_attn.in_proj_qkv,0.0002702387,0.05000,1.989 | |
| 33,linear_attn.in_proj_z,0.0001137513,0.05000,2.022 | |
| 33,linear_attn.out_proj,0.0000057096,0.05000,2.358 | |
| 33,mlp.gate_proj,0.0001562763,0.05000,2.871 | |
| 33,mlp.up_proj,0.0001783576,0.05000,2.878 | |
| 33,mlp.down_proj,0.0000078964,0.05000,9.727 | |
| 34,linear_attn.in_proj_qkv,0.0002623729,0.05000,2.051 | |
| 34,linear_attn.in_proj_z,0.0001201794,0.05000,2.045 | |
| 34,linear_attn.out_proj,0.0000058287,0.05000,2.418 | |
| 34,mlp.gate_proj,0.0001680708,0.05000,2.780 | |
| 34,mlp.up_proj,0.0001894957,0.05000,2.798 | |
| 34,mlp.down_proj,0.0000102192,0.05000,9.354 | |
| 35,self_attn.k_proj,0.0000276489,0.05000,3.962 | |
| 35,self_attn.v_proj,0.0000388686,0.05000,3.962 | |
| 35,self_attn.q_proj,0.0002702927,0.05000,3.983 | |
| 35,self_attn.o_proj,0.0000094047,0.05000,2.309 | |
| 35,mlp.gate_proj,0.0001789629,0.05000,2.753 | |
| 35,mlp.up_proj,0.0001953928,0.05000,2.763 | |
| 35,mlp.down_proj,0.0000115370,0.05000,9.672 | |
| 36,linear_attn.in_proj_qkv,0.0003655588,0.05000,2.060 | |
| 36,linear_attn.in_proj_z,0.0001855907,0.05000,1.984 | |
| 36,linear_attn.out_proj,0.0000057162,0.05000,2.465 | |
| 36,mlp.up_proj,0.0001925602,0.05000,2.874 | |
| 36,mlp.gate_proj,0.0001991262,0.05000,2.885 | |
| 36,mlp.down_proj,0.0000104420,0.05000,9.674 | |
| 37,linear_attn.in_proj_qkv,0.0003918287,0.05000,1.928 | |
| 37,linear_attn.in_proj_z,0.0002275343,0.05000,2.004 | |
| 37,linear_attn.out_proj,0.0000059617,0.05000,2.311 | |
| 37,mlp.gate_proj,0.0002017582,0.05000,2.795 | |
| 37,mlp.up_proj,0.0001917516,0.05000,2.853 | |
| 37,mlp.down_proj,0.0000098027,0.05000,9.710 | |
| 38,linear_attn.in_proj_qkv,0.0003525520,0.05000,2.006 | |
| 38,linear_attn.in_proj_z,0.0002091738,0.05000,1.879 | |
| 38,linear_attn.out_proj,0.0000046830,0.05000,2.345 | |
| 38,mlp.gate_proj,0.0002641791,0.05000,2.838 | |
| 38,mlp.up_proj,0.0002142234,0.05000,3.150 | |
| 38,mlp.down_proj,0.0000099394,0.05000,9.727 | |
| 39,self_attn.v_proj,0.0000394056,0.05000,4.145 | |
| 39,self_attn.k_proj,0.0000318873,0.05000,4.213 | |
| 39,self_attn.q_proj,0.0003112402,0.05000,4.256 | |
| 39,self_attn.o_proj,0.0000060011,0.05000,2.287 | |
| 39,mlp.gate_proj,0.0002393658,0.05000,2.810 | |
| 39,mlp.up_proj,0.0002080120,0.05000,2.840 | |
| 39,mlp.down_proj,0.0000092458,0.05000,9.717 | |
| 40,linear_attn.in_proj_qkv,0.0003310422,0.05000,2.307 | |
| 40,linear_attn.in_proj_z,0.0001864410,0.05000,1.976 | |
| 40,linear_attn.out_proj,0.0000053041,0.05000,2.429 | |
| 40,mlp.up_proj,0.0002204887,0.05000,2.837 | |
| 40,mlp.gate_proj,0.0002607984,0.05000,2.884 | |
| 40,mlp.down_proj,0.0000098098,0.05000,9.740 | |
| 41,linear_attn.in_proj_qkv,0.0003050378,0.05000,1.945 | |
| 41,linear_attn.in_proj_z,0.0001696298,0.05000,1.996 | |
| 41,linear_attn.out_proj,0.0000046962,0.05000,2.688 | |
| 41,mlp.up_proj,0.0002130802,0.05000,2.855 | |
| 41,mlp.gate_proj,0.0002247281,0.05000,2.862 | |
| 41,mlp.down_proj,0.0000096282,0.05000,9.423 | |
| 42,linear_attn.in_proj_qkv,0.0002849892,0.05000,2.075 | |
| 42,linear_attn.in_proj_z,0.0001530167,0.05000,1.877 | |
| 42,linear_attn.out_proj,0.0000052826,0.05000,2.569 | |
| 42,mlp.gate_proj,0.0002039032,0.05000,2.780 | |
| 42,mlp.up_proj,0.0002093354,0.05000,2.879 | |
| 42,mlp.down_proj,0.0000104336,0.05000,9.630 | |
| 43,self_attn.k_proj,0.0000311214,0.05000,3.924 | |
| 43,self_attn.v_proj,0.0000381165,0.05000,4.046 | |
| 43,self_attn.q_proj,0.0002958142,0.05000,4.107 | |
| 43,self_attn.o_proj,0.0000087401,0.05000,2.359 | |
| 43,mlp.up_proj,0.0002176778,0.05000,2.853 | |
| 43,mlp.gate_proj,0.0002034712,0.05000,2.874 | |
| 43,mlp.down_proj,0.0000114969,0.05000,10.142 | |
| 44,linear_attn.in_proj_qkv,0.0002739369,0.05000,2.060 | |
| 44,linear_attn.in_proj_z,0.0001355589,0.05000,1.968 | |
| 44,linear_attn.out_proj,0.0000074509,0.05000,2.427 | |
| 44,mlp.up_proj,0.0002208027,0.05000,2.832 | |
| 44,mlp.gate_proj,0.0002033508,0.05000,2.878 | |
| 44,mlp.down_proj,0.0000129796,0.05000,9.866 | |
| 45,linear_attn.in_proj_qkv,0.0002792265,0.05000,2.067 | |
| 45,linear_attn.in_proj_z,0.0001384729,0.05000,2.072 | |
| 45,linear_attn.out_proj,0.0000065496,0.05000,2.394 | |
| 45,mlp.up_proj,0.0002234423,0.05000,2.918 | |
| 45,mlp.gate_proj,0.0002046250,0.05000,2.939 | |
| 45,mlp.down_proj,0.0000134076,0.05000,9.396 | |
| 46,linear_attn.in_proj_qkv,0.0002909530,0.05000,2.012 | |
| 46,linear_attn.in_proj_z,0.0001456825,0.05000,2.002 | |
| 46,linear_attn.out_proj,0.0000095057,0.05000,2.464 | |
| 46,mlp.gate_proj,0.0002010068,0.05000,2.793 | |
| 46,mlp.up_proj,0.0002256891,0.05000,2.816 | |
| 46,mlp.down_proj,0.0000143541,0.05000,9.666 | |
| 47,self_attn.v_proj,0.0000506427,0.05000,3.942 | |
| 47,self_attn.q_proj,0.0003168942,0.05000,4.071 | |
| 47,self_attn.k_proj,0.0000321328,0.05000,4.077 | |
| 47,self_attn.o_proj,0.0000101120,0.05000,2.398 | |
| 47,mlp.gate_proj,0.0002227599,0.05000,2.660 | |
| 47,mlp.up_proj,0.0002490375,0.05000,2.815 | |
| 47,mlp.down_proj,0.0000175623,0.05000,9.421 | |
| 48,linear_attn.in_proj_qkv,0.0003186905,0.05000,1.999 | |
| 48,linear_attn.in_proj_z,0.0001540790,0.05000,1.970 | |
| 48,linear_attn.out_proj,0.0000110738,0.05000,2.431 | |
| 48,mlp.gate_proj,0.0002386876,0.05000,2.867 | |
| 48,mlp.up_proj,0.0002675655,0.05000,2.881 | |
| 48,mlp.down_proj,0.0000212672,0.05000,9.621 | |
| 49,linear_attn.in_proj_qkv,0.0003820718,0.05000,2.009 | |
| 49,linear_attn.in_proj_z,0.0001642076,0.05000,1.910 | |
| 49,linear_attn.out_proj,0.0000149738,0.05000,2.275 | |
| 49,mlp.gate_proj,0.0002584349,0.05000,2.741 | |
| 49,mlp.up_proj,0.0002866655,0.05000,2.790 | |
| 49,mlp.down_proj,0.0000272903,0.05000,9.620 | |
| 50,linear_attn.in_proj_qkv,0.0003893767,0.05000,2.039 | |
| 50,linear_attn.in_proj_z,0.0001826149,0.05000,1.886 | |
| 50,linear_attn.out_proj,0.0000220006,0.05000,2.407 | |
| 50,mlp.up_proj,0.0003265009,0.05000,2.806 | |
| 50,mlp.gate_proj,0.0003083456,0.05000,2.852 | |
| 50,mlp.down_proj,0.0000455937,0.05000,9.704 | |
| 51,self_attn.k_proj,0.0000418068,0.05000,3.974 | |
| 51,self_attn.q_proj,0.0004121897,0.05000,4.004 | |
| 51,self_attn.v_proj,0.0000925749,0.05000,4.031 | |
| 51,self_attn.o_proj,0.0000293543,0.05000,2.477 | |
| 51,mlp.gate_proj,0.0003434407,0.05000,2.824 | |
| 51,mlp.up_proj,0.0003731706,0.05000,2.881 | |
| 51,mlp.down_proj,0.0000584166,0.05000,9.791 | |
| 52,linear_attn.in_proj_qkv,0.0005982841,0.05000,2.058 | |
| 52,linear_attn.in_proj_z,0.0002757637,0.05000,2.004 | |
| 52,linear_attn.out_proj,0.0000300584,0.05000,2.357 | |
| 52,mlp.gate_proj,0.0004216733,0.05000,2.768 | |
| 52,mlp.up_proj,0.0004031492,0.05000,3.058 | |
| 52,mlp.down_proj,0.0000588965,0.05000,9.417 | |
| 53,linear_attn.in_proj_qkv,0.0005947312,0.05000,1.999 | |
| 53,linear_attn.in_proj_z,0.0002927151,0.05000,2.049 | |
| 53,linear_attn.out_proj,0.0000332334,0.05000,2.384 | |
| 53,mlp.up_proj,0.0004268191,0.05000,2.878 | |
| 53,mlp.gate_proj,0.0004694973,0.05000,2.915 | |
| 53,mlp.down_proj,0.0000641496,0.05000,9.748 | |
| 54,linear_attn.in_proj_qkv,0.0005754633,0.05000,2.075 | |
| 54,linear_attn.in_proj_z,0.0003124722,0.05000,1.922 | |
| 54,linear_attn.out_proj,0.0000333667,0.05000,2.403 | |
| 54,mlp.up_proj,0.0005059157,0.05000,2.791 | |
| 54,mlp.gate_proj,0.0005929848,0.05000,2.831 | |
| 54,mlp.down_proj,0.0000913851,0.05000,9.517 | |
| 55,self_attn.q_proj,0.0005465929,0.05000,3.951 | |
| 55,self_attn.v_proj,0.0001239293,0.05000,3.981 | |
| 55,self_attn.k_proj,0.0000579751,0.05000,4.000 | |
| 55,self_attn.o_proj,0.0000298551,0.05000,2.327 | |
| 55,mlp.gate_proj,0.0005700416,0.05000,2.823 | |
| 55,mlp.up_proj,0.0005092237,0.05000,2.834 | |
| 55,mlp.down_proj,0.0000769449,0.05000,9.598 | |
| 56,linear_attn.in_proj_qkv,0.0006624743,0.05000,1.882 | |
| 56,linear_attn.in_proj_z,0.0003330500,0.05000,1.871 | |
| 56,linear_attn.out_proj,0.0000457990,0.05000,2.365 | |
| 56,mlp.gate_proj,0.0006302628,0.05000,2.915 | |
| 56,mlp.up_proj,0.0005478271,0.05000,3.194 | |
| 56,mlp.down_proj,0.0000767774,0.05000,9.813 | |
| 57,linear_attn.in_proj_qkv,0.0007052206,0.05000,2.025 | |
| 57,linear_attn.in_proj_z,0.0003543249,0.05000,2.000 | |
| 57,linear_attn.out_proj,0.0000407313,0.05000,2.426 | |
| 57,mlp.up_proj,0.0006268600,0.05000,2.788 | |
| 57,mlp.gate_proj,0.0007333108,0.05000,2.868 | |
| 57,mlp.down_proj,0.0000806746,0.05000,9.734 | |
| 58,linear_attn.in_proj_qkv,0.0006764381,0.05000,2.249 | |
| 58,linear_attn.in_proj_z,0.0003702752,0.05000,1.827 | |
| 58,linear_attn.out_proj,0.0000439231,0.05000,2.325 | |
| 58,mlp.up_proj,0.0007224930,0.05000,2.771 | |
| 58,mlp.gate_proj,0.0008458127,0.05000,2.851 | |
| 58,mlp.down_proj,0.0001073471,0.05000,9.706 | |
| 59,self_attn.v_proj,0.0002534219,0.05000,3.880 | |
| 59,self_attn.k_proj,0.0000720708,0.05000,3.934 | |
| 59,self_attn.q_proj,0.0006471389,0.05000,4.061 | |
| 59,self_attn.o_proj,0.0000865519,0.05000,2.479 | |
| 59,mlp.up_proj,0.0007308760,0.05000,2.750 | |
| 59,mlp.gate_proj,0.0008335251,0.05000,2.801 | |
| 59,mlp.down_proj,0.0001261923,0.05000,9.394 | |
| 60,linear_attn.in_proj_qkv,0.0008105025,0.05000,2.106 | |
| 60,linear_attn.in_proj_z,0.0003880060,0.05000,2.010 | |
| 60,linear_attn.out_proj,0.0001007381,0.05000,2.189 | |
| 60,mlp.gate_proj,0.0008283832,0.05000,2.740 | |
| 60,mlp.up_proj,0.0007406551,0.05000,2.780 | |
| 60,mlp.down_proj,0.0001567032,0.05000,9.783 | |
| 61,linear_attn.in_proj_qkv,0.0006010910,0.05000,1.947 | |
| 61,linear_attn.in_proj_z,0.0003464860,0.05000,1.794 | |
| 61,linear_attn.out_proj,0.0000901399,0.05000,2.424 | |
| 61,mlp.up_proj,0.0007882093,0.05000,2.968 | |
| 61,mlp.gate_proj,0.0008725039,0.05000,2.987 | |
| 61,mlp.down_proj,0.0002059536,0.05000,9.399 | |
| 62,linear_attn.in_proj_qkv,0.0006572308,0.05000,2.043 | |
| 62,linear_attn.in_proj_z,0.0003461558,0.05000,2.032 | |
| 62,linear_attn.out_proj,0.0002392242,0.05000,2.488 | |
| 62,mlp.up_proj,0.0007274505,0.05000,2.755 | |
| 62,mlp.gate_proj,0.0008007132,0.05000,2.820 | |
| 62,mlp.down_proj,0.0003095795,0.05000,9.611 | |
| 63,self_attn.v_proj,0.0002023079,0.05000,3.974 | |
| 63,self_attn.k_proj,0.0000646610,0.05000,3.993 | |
| 63,self_attn.q_proj,0.0005759815,0.05000,4.071 | |
| 63,self_attn.o_proj,0.0002338829,0.05000,2.411 | |
| 63,mlp.gate_proj,0.0006730167,0.05000,2.771 | |
| 63,mlp.up_proj,0.0005922088,0.05000,2.782 | |
| 63,mlp.down_proj,0.0006151453,0.05000,9.336 | |