issues
search
sunzeyeah
/
RLHF
Implementation of Chinese ChatGPT
282
stars
36
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Pangu 2.6b 启动失败。
#25
Liufeiran123
opened
6 months ago
1
pretrain_data_v1.jsonl 文件在哪里
#24
Liufeiran123
closed
5 months ago
5
基于ChatGLM2的RLHF训练问题
#23
UltraZeroyH
opened
10 months ago
1
有这个推理的代码吗
#22
ww0o0
opened
11 months ago
0
后续会考虑RLHF平替方案的集成么
#21
skykiseki
closed
11 months ago
2
请问为什么我在基于Chatglm-6b模型执行train_sft.sh训练时，启用deepspeed没有效果
#20
IconShan
closed
4 months ago
2
请问支持lora方式吗
#19
70557dzqc
closed
1 year ago
1
chatglm+RLHF
#18
MAJIN123
closed
1 year ago
1
关于取最后一个token作为reward分数的方式
#17
Bo396543018
closed
1 year ago
2
reward推理问题
#16
ItGirls
closed
1 year ago
5
train_rlhf-trlx.py代码问题
#15
taofennanhai
closed
1 year ago
3
RLHF第三步进行训练时，在modeling_glm.py如下代码位置会显示“size mismatch, got 1028, 1028x1024,0”
#14
IconShan
closed
1 year ago
4
rlhf deepspeed和trlx能否支持 sft chatglm 6b
#13
GUORUIWANG
closed
1 year ago
13
reward model的实现问题
#12
DamonYangyang
closed
1 year ago
1
用GLM-10B-chinese训练RLHF过程，有没有模型并行的方式？
#11
taofennanhai
closed
1 year ago
4
用chatGLM-6B训RW的时候loss不收敛
#10
GUORUIWANG
closed
1 year ago
7
使用LoRA的GLM-10B-chinese模型是如何保存的
#9
taofennanhai
closed
1 year ago
3
No module named 'transformers_modules.sunzeyeah.pangu-2'
#8
MRKINKI
closed
1 year ago
4
请教一下有考虑加IDEA-CCNL/Wenzhong2.0-GPT2-3.5B-chinese这个模型做评测比较吗
#7
MRKINKI
closed
1 year ago
1
deepspeed速度
#6
superqing001
closed
1 year ago
0
RLHF相关问题
#5
taofennanhai
closed
1 year ago
4
有对比不加RLHF和加入RLHF的效果吗
#4
macheng6
closed
1 year ago
1
训练相关
#3
macheng6
closed
1 year ago
2
为什么训练的时候要加入<sep> token？
#2
Nipi64310
closed
1 year ago
3
加入RW后模型的效果
#1
yxk9810
closed
1 year ago
2