issues
search
OpenLLMAI
/
OpenRLHF
An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & Mixtral)
https://openrlhf.readthedocs.io/
Apache License 2.0
1.71k
stars
160
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
PPO采用zero 3 stage后产生time out error
#293
victorShawFan
opened
1 month ago
3
启用PPO Ray后无响应
#292
victorShawFan
closed
1 month ago
3
RLHF for classification tasks
#291
vinodrajendran001
closed
1 day ago
2
HTTPError when running train_ppo_llama_ray.sh
#290
Zeyuan-Liu
closed
1 day ago
5
[question] long context for single model ppo training
#289
yananchen1989
closed
1 month ago
1
RM training loss becomes NAN when finish the first training step.
#288
lixsh6
opened
1 month ago
1
PPO训练之后模型拒绝回答
#287
burger-pb
closed
1 day ago
3
Vllm0.42 + Lora configs
#286
hijkzzz
closed
1 month ago
0
Custom ExperienceMaker
#285
mgerstgrasser
opened
2 months ago
4
when import requests, class NewLineFormatter(logging.Formatter): AttributeError: partially initialized module 'logging' has no attribute 'Formatter' (most likely due to a circular import)
#284
catqaq
opened
2 months ago
0
fix vLLM v0.4.1
#283
hijkzzz
closed
2 months ago
3
Update NGC and vllm version.
#282
THINK2TRY
closed
2 months ago
2
Revert "vllm 0.4.1 compatibility (#278)"
#281
hijkzzz
closed
2 months ago
0
fix typos in train_ppo_ray.py
#280
mickelliu
closed
2 months ago
1
fix typos in train_ppo_ray.py
#279
mickelliu
closed
2 months ago
0
vllm 0.4.1 compatibility
#278
mgerstgrasser
closed
2 months ago
6
内存超出问题
#277
burger-pb
closed
1 day ago
3
upgrade transformer/deespeed and sync when colocate models
#276
hijkzzz
closed
2 months ago
1
CUDA out of memory when i run train_ppo_llama_ray.sh on 4 RTX 4090(24G)
#275
libowen424
closed
2 months ago
2
AssertionError: mismatch size output_state_dict(148) and state_dict(149) sft training
#274
qwenzo
closed
2 months ago
3
reward model数据集问题
#273
burger-pb
closed
1 day ago
3
PPO training configuration for train_ppo_llama.sh
#272
MurrayTom
closed
1 day ago
1
NCCL Broad cast error after first actor fit
#271
karthik-nexusflow
closed
2 months ago
17
Issue with models not using `position_ids`
#270
kfertakis
opened
2 months ago
1
The configuration for Llama-7b on 4 RTX4090
#269
LinkyLiu
opened
2 months ago
5
Inconsistent python version dependency
#268
snailrowen1337
closed
2 months ago
1
add test pipeline: use small LLM and small data
#267
catqaq
opened
2 months ago
0
Documentation for using Kuberay
#266
karthik-nexusflow
opened
2 months ago
4
vllm / actor broadcast error in multinode training
#265
karthik-nexusflow
closed
2 months ago
28
add Knowledge Distillation
#264
ifromeast
closed
2 months ago
3
[Baseline] LLaMA2-7B RLHF training curves
#263
hijkzzz
opened
2 months ago
2
How long does single LLM's tunning reuqired?
#262
alphahumancoder
opened
2 months ago
3
新添中文的READDME.md文件
#261
khazic
closed
2 months ago
0
How to get score for a single response from a trained RM
#260
UbeCc
closed
2 months ago
1
use custom datasets and cache_dir
#259
UbeCc
closed
2 months ago
4
debugging with ray
#258
mickel-liu
closed
3 months ago
2
[pre-commit.ci] pre-commit suggestions
#257
pre-commit-ci[bot]
closed
3 months ago
0
Is save checkpoint not yet supported for ppo ray trainer?
#256
mickel-liu
opened
3 months ago
5
how to train in fp16?
#255
dshnightmare
closed
3 months ago
1
Hardware requirement
#254
ridiculouz
closed
3 months ago
6
Support ORPO
#253
paulcx
opened
3 months ago
1
this repo's hack of rope embedding accepts different input than transformers
#252
babu111
closed
3 months ago
3
[For your information] Ways to build environment and run openrlhf codes on a slurm cluster
#251
glorgao
closed
2 months ago
2
add perf and benchmark scripts
#249
wuxibin89
closed
1 month ago
3
fix num_training_steps when micro rollout and train size are not equal
#248
wuxibin89
closed
3 months ago
0
Fixed error due to 'margin' variable type being list in rm_trainer.py
#247
StwayneXG
closed
3 months ago
0
Unexpected long actor_time when train_ppo_ray
#246
LSC527
opened
3 months ago
9
enable_ema cause runtime error when running train_ppo_llama.sh
#245
dshnightmare
opened
3 months ago
6
Update requirements.txt
#244
kfertakis
closed
3 months ago
0
Issues with bulding OpenRLHF locally
#243
kfertakis
closed
2 months ago
3
Previous
Next