Ant0082
Ant0082
Is it fine-tuned on task-specific datasets using OpenDelta ？
attach config.json { "session_token":"xxxxx" }
1. where to get a proxy? export https_proxy=http://127.0.0.1:7890 http_proxy=http://127.0.0.1:7890 all_proxy=socks5://127.0.0.1:7890 { "proxy":"http://127.0.0.1:7890" } Is this the way to fill in? 2. Can you give a sample config.json file?
> For finetuning, the optimize states consume a lot of memory. You can enable ZeRO-Offload (https://www.deepspeed.ai/tutorials/zero-offload/) to offload the optimizer states to the CPU memory. By default, we already enable...
> 想问一下，如果设置了"cpu_offload": true，需要大概多少的显存呢？8*16G的V100是否能够跑起来呢 设置 stage=2，加300G内存，8*16G应该可以跑起来 stage=2 300G 4*40G(实际占用～33G) 可以跑起来。
感谢您的回复。
@zhaoxudong-inspur 您好，API中的dialog参数量级是多少？
Hello, has this issue been resolved?