zll0000

Results 6 issues of zll0000

[2022-06-29 15:38:16] [config] ulr-keys-vectors: "" [2022-06-29 15:38:16] [config] ulr-query-vectors: "" [2022-06-29 15:38:16] [config] ulr-softmax-temperature: 1 [2022-06-29 15:38:16] [config] ulr-trainable-transformation: false [2022-06-29 15:38:16] [config] version: v1.8.40 24df8f1c 2020-02-15 13:47:18 -0800 [2022-06-29...

bug

A/opt/conda/conda-bld/pytorch_1634272068694/work/aten/src/ATen/native/cuda/IndexKernel.cu:93: operator(): block: [0,0,0], thread: [0,0,0] Assertion `index >= -sizes[i] && index < sizes[i] && "index out of bounds"` failed. /opt/conda/conda-bld/pytorch_1634272068694/work/aten/src/ATen/native/cuda/IndexKernel.cu:93: operator(): block: [0,0,0], thread: [1,0,0] Assertion `index >= -sizes[i]...

evalscope perf --url 'http://127.0.0.1:65000/v1/chat/completions' --parallel 128 --model 'Qwen2-72B-Instruct' --log-every-n-query 10 --read-timeout=120 --dataset-path './data/open_qa.jsonl' -n 1 --max-prompt-length 128000 --api openai --stream --stop '' --dataset openqa --debug --wandb-api-key 'my_api_key' 运训完后卡住 s":[{"index":0,"delta":{"content":"风格和"},"logprobs":null,"finish_reason":null}],"usage":null} 2024-08-07...

vllm 部署的服务 为啥请求不通 evalscope perf --url 'our_url/v1/completions' --parallel 128 --model 'Qwen2-72B-Instruct' --log-every-n-query 10 --read-timeout=120 --dataset-path './data/open_qa.jsonl' -n 1 --max-prompt-length 128000 --api openai --stream --stop '' --dataset openqa --debug 在本地机器上 利用vllm...

### Reminder - [x] I have read the above rules and searched the existing issues. ### System Info 利用mllm_audio_demo 数据测试微调 miniccpm-o-2.6 #####yaml: ### model model_name_or_path: /workdir/model/minicpm-o-2-6 trust_remote_code: true ### method...

bug
pending