Token Issue 230 Facicochinese Vicuna Github
Bug in token of Chinese[Bug] · Issue #2482 · coqui-ai/TTS · GitHub
why choosing vicuna as the tokenizer · Issue #53 · locuslab/wanda · GitHub
训练模型没有保存token信息 · Issue #230 · Facico/Chinese-Vicuna · GitHub
output 生成内容为乱码 · Issue #2 · Facico/Chinese-Vicuna · GitHub
微调之后加载权重发现输出停不下来 · Issue #140 · Facico/Chinese-Vicuna · GitHub
Tokenizer issue with Vicuna V1.1, EOS, BOS tokens seem to be blank ...
训练的3.0轮是够的么？ · Issue #136 · Facico/Chinese-Vicuna · GitHub
请问用4090单卡是否可以lora微调13b模型？ · Issue #146 · Facico/Chinese-Vicuna · GitHub
微调之后加载权重发现输出停不下来 · Issue #140 · Facico/Chinese-Vicuna · GitHub
请教下单机多卡训练的卡死问题 · Issue #49 · Facico/Chinese-Vicuna · GitHub
Advertisement Space (300x250)
数据集无法下载 · Issue #67 · Facico/Chinese-Vicuna · GitHub
这是什么情况，后面的回答都是回答的第一个问题？ · Issue #119 · Facico/Chinese-Vicuna · GitHub
Vicuna template may cause the tokenization mismatch warning · Issue ...
transformers和pydantic问题 · Issue #238 · Facico/Chinese-Vicuna · GitHub
单机多卡一直报超时错误，请教下大佬有没有啥解决的办法啊 · Issue #53 · Facico/Chinese-Vicuna · GitHub
vicuna-7b-v1.5 err · Issue #2312 · lm-sys/FastChat · GitHub
这是什么情况，后面的回答都是回答的第一个问题？ · Issue #119 · Facico/Chinese-Vicuna · GitHub
关于多卡训练的问题 · Issue #96 · Facico/Chinese-Vicuna · GitHub
continue finetune vicuna using `--resume_from_checkpoint` · Issue #238 ...
结果不是很好 · Issue #198 · Facico/Chinese-Vicuna · GitHub
Advertisement Space (336x280)
微调之后加载权重发现输出停不下来 · Issue #140 · Facico/Chinese-Vicuna · GitHub
How to run vicuna on a CPU with 16GB of RAM? · Issue #129 · Facico ...
为什么fine-tune过程中loss会忽大忽小呢？ · Issue #39 · Facico/Chinese-Vicuna · GitHub
为什么fine-tune过程中loss会忽大忽小呢？ · Issue #39 · Facico/Chinese-Vicuna · GitHub
CUBLAS_STATUS_INTERNAL_ERROR for OPT-13b when input token length > 230 ...
运行generate.py没有反应 · Issue #101 · Facico/Chinese-Vicuna · GitHub
训练的3.0轮是够的么？ · Issue #136 · Facico/Chinese-Vicuna · GitHub
readme 的环境配置指导文件是否存在错误 ？ · Issue #42 · Facico/Chinese-Vicuna · GitHub
readme 的环境配置指导文件是否存在错误 ？ · Issue #42 · Facico/Chinese-Vicuna · GitHub
Unable to run other models - e.g. Vicuna 7b · Issue #50 · ParisNeo ...
Advertisement Space (336x280)
13b的结果,很奇怪 · Issue #15 · dandelionsllm/pandallm · GitHub
为什么我训练的lora无论是启动、还是推理的速度都很慢 · Issue #181 · Facico/Chinese-Vicuna · GitHub
Text generation early stop problem with Vicuna 33B v1.3 · Issue #1821 ...
TheBloke/stable-vicuna-13B-GPTQ · Issue starting the webui
generate和interaction都无法停止，直到达到max_tokens限制才会停止 · Issue #59 · Facico ...
Chinese-Vicuna/generate.py at master · Facico/Chinese-Vicuna · GitHub