issues
search
CVI-SZU
/
Linly
Chinese-LLaMA 1&2、Chinese-Falcon 基础模型；ChatFlow中文对话模型；中文OpenLLaMA模型；NLP预训练/指令微调数据集
3.03k
stars
234
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
llama3增量预训练冻结哪些层训练哪些层效果比较好？
#137
CanvaChen
closed
5 months ago
0
请问有没有性别年龄检测模型？
#136
broadcast98
opened
6 months ago
0
请问70B的模型要如何使用，抱脸上的模型看着文件和其他模型不一样
#135
afezeriaWrnbbmm
opened
9 months ago
0
pretrain.py的示例似乎有点错误
#134
xinghudamowang
opened
10 months ago
0
在线地址无法使用
#133
zzSleep
opened
11 months ago
0
服务器最低配置要求是什么？
#132
jqs1124
opened
12 months ago
0
有人有pile的数据集吗？22个来源，825G的那个版本
#131
AI-Study-Han
opened
1 year ago
0
问下大佬们有没有训练3B的打算？场景需要时延不能太高
#130
zjuzhfbloodz
opened
1 year ago
0
readme上的加群二维码过期了
#129
potong
opened
1 year ago
0
Chinese-LLaMA-2-13B-hf样本模板prompt到底是什么样的？
#128
sunzhaowei
opened
1 year ago
0
优化readme内容 (#1)
#127
Smilencelsy
closed
1 year ago
0
请问，deepspeed 微调时，CPU的内存需要多大？
#126
Batmana
opened
1 year ago
1
关于Chinese-LLaMA-2-13B (hf格式)
#125
sun1092469590
opened
1 year ago
0
Please clarify the License for Chinese-LLaMA-2
#124
JayLiangs
opened
1 year ago
1
微信满员了，请重新上传新的微信图片 我可以免费做管理员
#123
ArtificialZeng
closed
1 year ago
3
多轮对话问问题之后直接报错
#122
caowenhero
opened
1 year ago
0
python3 llama_server.py结果乱码
#121
caowenhero
opened
1 year ago
0
ChatFlow-13B.bin只有136字节
#120
NewEricWang
opened
1 year ago
1
huggingface上openllama-13b的模型大小为26.4G,转换为huggingface那种模型格式之后模型大小为24.7G，这也就是大概是以fp16或者是bf16保存的模型
#119
belle9217
opened
1 year ago
0
Are the tokenizer.model the same with the one in llama-7b?
#118
treya-lin
opened
1 year ago
0
Chinese-LLaMA-33B在多少块gpu上训了多长时间？
#117
JingxinLee
opened
1 year ago
0
是否考虑通过位置插值来扩展大语言模型的上下文窗口 ，将上下文窗口提升至32K
#116
xfg0913
opened
1 year ago
0
请问在指令微调时损失函数与预训练有什么区别吗
#115
dazhaxie0526
opened
1 year ago
0
open-llama13B做推理时，结果是英文
#114
yating0823
opened
1 year ago
0
falcon的使用中文预料进行增量训练
#113
fengstar7827
opened
1 year ago
2
额，是我用错了吗?简单推理都不行吗
#112
Mousaic
opened
1 year ago
1
Multi machine pre-training hung
#111
BUPTAnderson
opened
1 year ago
1
请问有中文falcon的下载地址嘛
#110
AlexXx-Wu
opened
1 year ago
2
Wrong argments
#109
jeffchy
opened
1 year ago
0
chatflow模型推理的时候，prompt需要加类似“human: {query}\n assistant: \n” 前后缀嘛？
#108
jeinlee1991
opened
1 year ago
0
openllama 13b base model生成内容比较奇怪
#107
lucasjinreal
opened
1 year ago
0
关于openllama的两个相关问题
#106
lucasjinreal
opened
1 year ago
1
请问是否有增量预训练的基础模型13B的评测结果？
#105
caihaunqai
opened
1 year ago
0
使用openllama13B + openmodel进行推理时，结果都是数字？这个需要做其他操作？
#104
suhaibo1
opened
1 year ago
1
增量预训练的时候报错exits with return code = -9 ，单卡80G显存的A100
#103
pydaxing
opened
1 year ago
2
HF在线崩溃了
#102
Batmana
opened
1 year ago
1
Pretraining corpus formatting
#101
treya-lin
opened
1 year ago
0
7b模型性能和billa对比
#100
lucasjinreal
opened
1 year ago
0
readme上的加群二维码过期了
#99
aihaidong
opened
1 year ago
4
如何cite？
#98
hackerchenzhuo
opened
1 year ago
1
请问大佬65B的模型何时能够放出
#97
Expert68
opened
1 year ago
0
关于33B模型预训练语料长度
#96
minlik
opened
1 year ago
2
请问openllama 13b怎么转成HF格式
#95
lin1490188
opened
1 year ago
0
请问OpenLLaMA-13B在转换为hf模型时，convert_llama_from_tencentpretrain_to_hf.py直接复制了词表tokenizer.model，open_llama.model没有用到，是正常的吗？
#94
chk4991
opened
1 year ago
1
关于平行语料的预处理
#93
lyy-zz
opened
1 year ago
3
openllama性能评估
#92
enbacoo
opened
1 year ago
0
请问模型在tencentpretrain框架下预训练时选择的是bpe tokenizer吗？是否有对应的预训练的merge.txt呢？
#91
yyqi17
opened
1 year ago
0
Chinese-LLaMA-33B (hf格式)的模型如何部署，进行推理？
#90
xfg0913
opened
1 year ago
1
33b Huggingface 格式怎么转成TencentPretrain 格式
#89
lyy-zz
closed
1 year ago
2
Is it possible to support OPT models
#88
treya-lin
opened
1 year ago
1
Next