issues
search
ymcui
/
Chinese-LLaMA-Alpaca
中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)
https://github.com/ymcui/Chinese-LLaMA-Alpaca/wiki
Apache License 2.0
18.23k
stars
1.86k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
微调模型数据填充问题
#768
xzqxnet0990
closed
1 year ago
4
v5.0: Release Pro model series and Plus-33B models
#766
ymcui
closed
1 year ago
0
请问微调指令数据很长该怎么设置参数
#764
TheLolita
closed
1 year ago
2
Api模式能提供流式响应模式吗?
#762
guyuexue
closed
1 year ago
1
推理一直报错
#761
DRZJ1
closed
1 year ago
3
Update inference_hf.py for repetition_penalty
#760
ymcui
closed
1 year ago
0
训练中报错Unable to concatenate an empty list of datasets.main()
#759
DRZJ1
closed
1 year ago
6
OPEN_AI API调用速度很慢
#758
Cliffe97
closed
1 year ago
2
关于33B的plus版
#757
nuoma
closed
1 year ago
2
Adding load_in_8bit to inference_hf.py
#756
airaria
closed
1 year ago
0
Adding --load_in_8bit option to inference_hf.py
#755
airaria
closed
1 year ago
0
Alpaca-plus-7b微调时报错
#754
TCHSDUFH
closed
1 year ago
2
lora权重加载问题
#751
Yuang-Deng
closed
1 year ago
2
Chinese-Alpaca-Plus-7B 预训练报错
#750
mazhai
closed
1 year ago
2
Why build_dataset.py combine source = "instruction + input + output" ==> overfit immediately ?
#749
thusinh1969
closed
1 year ago
3
chinese_llama_plus_lora_7b中的tokenizer, vocab size不正确。
#748
yeli7068
closed
1 year ago
2
执行 inference_hf.py 时提示缺少文件
#747
XXXG00W0
closed
1 year ago
12
运行ingest.py发生错误
#746
handsomexiaoyi
closed
1 year ago
2
关于其他参数量模型的训练
#744
wuhuanon
closed
1 year ago
2
Add patches for memory_efficient_attention and NTK scaling
#743
airaria
closed
1 year ago
1
run_pt增量预训练，中断后重新训练，失败
#742
smartparrot
closed
1 year ago
1
Chinese-Alpaca-Plus-7B 预训练后回答效果不理想
#739
zhaomeng0113
closed
1 year ago
7
OSError: ziqingyang/chinese-llama-plus-lora-7b does not appear to have a file named config.json.
#738
tanshuai
closed
1 year ago
2
预训练后合并lora，之后load模型出现词表大小不对的问题
#737
Double-bear
closed
1 year ago
2
使用Transformers推理，最新的Transformers已经没有scripts/inference_hf.py脚本
#736
aawuj
closed
1 year ago
3
run_clm_sft_with_peft.py脚本是不是不支持shareGPT那种形式的多轮数据训练？
#735
xyfZzz
closed
1 year ago
4
RuntimeError: Function MmBackward0 returned an invalid gradient at index 1 - expected device meta but got cuda:0
#734
wangjvjie
closed
1 year ago
0
ValueError The vocab size of the tokenizer must be 49954, but found 49953
#733
wangjvjie
closed
1 year ago
1
继续训练Chinese-Alpaca模型的LoRA权重，新的lora与哪个模型合并呢
#731
Agreewithu
closed
1 year ago
7
lora预训练13B模型需要多大内存的GPU。单机双卡 2*24GB 会爆显
#730
wangxigui
closed
1 year ago
11
什么意思，这个图我真的绷不住了
#729
CXLiang123
closed
1 year ago
4
在加入lora时，提前没有做prepare_model_for_kbit_training
#728
guijuzhejiang
closed
1 year ago
1
继续预训练loss训飞
#727
HalcyonLiang
closed
1 year ago
9
Add news of Visual-Chinese-LLaMA-Alpaca
#726
airaria
closed
1 year ago
0
关于训练的基底模型
#725
wuhuanon
closed
1 year ago
3
我在mac上合并的模型，然后拷贝到win/linux上使用，性能会有影响么？
#723
stoneLee81
closed
1 year ago
1
一机多卡执行训练报错，torchrun 的 --nproc_per_node 配置`2`时正常，配置为大于`2`的数值后报错
#722
shibingli
closed
1 year ago
5
UserWarning: None of the inputs have requires_grad=True. Gradients will be None
#721
ljch2018
closed
1 year ago
3
预训练全量参数报错
#720
Double-bear
closed
1 year ago
0
Update banner path, change default decoding values for Gradio demo
#719
ymcui
closed
1 year ago
0
关于gradio_demo里prompt格式问题
#716
zeng9t
closed
1 year ago
5
chinese_sp.model是如何训练的，是否能给出详细步骤及代码实现
#713
zemu121
closed
1 year ago
6
langchain示例未正確輸出
#712
cyc00518
closed
1 year ago
2
lora训练保存的adapter_model.bin很小，只有443字节
#711
guijuzhejiang
closed
1 year ago
16
模型sft训练过程中进度条卡住一直不动，也不报错
#710
Qmymy
closed
1 year ago
2
13B模型合并SHA256不一致，指令精调报错
#709
slxy-hub
closed
1 year ago
2
请问在中文LLaMa进行sft的数据量是多少呀，想复现一下，数据因该是5w条的json，但是训练多少个epoch或者token数呀，我看介绍是指令4M ，这个没太理解是怎么算的。
#708
sixgold993
closed
1 year ago
6
Fix unexpected slow down in gradio web demo
#707
GoGoJoestar
closed
1 year ago
2
Colab 最后量化为4bit时报错，4096*49954不能被256整除
#706
Ziffer-byakuya
closed
1 year ago
2
Extend context size without fine-tuning
#705
airaria
closed
1 year ago
4
Previous
Next