-
Hi, I following the script to train the bloom model for my own dataset. However, I found that it saved the model differently as compared to other models such as OPT. The screenshot below shows the sav…
-
### 🚀 The feature, motivation, and pitch
Thank you for your great work.
I used scipts in summarize_rlhf train own ppo model(gpt-2) and tried to do inference.
After converting trained model to bin f…
-
### Week 1 - Get to know the community
- [X] Join the communication channels
- [X] Open a GitHub issue (this one!)
- [X] Install the Ersilia Model Hub and test the simplest model
- [X] Write a motiva…
-
Hello, trying to figure out why my h2ogpt doesn't use my GPU at all. Figured that something has to be wrong with bitsandbytes, since it says it was compiled without GPU support. I made everything work…
-
Hello,
I have successfully run the code summarize_rlhf with small SFT and RM models (bloom1b). However, when I try to run the larger model (7B), **the timeout error is raised,** which is a similar …
-
Hi, Kev
It's glad to know your work about DTQN.
I am very curious about why the work of combine Transformer and DQN is very small ,and this two technology is emit very early.
Because I thought ther…
-
# 每日安全资讯（2023-06-01）
- Files ≈ Packet Storm
- [ ] [Qualcomm Adreno/KGSL Data Leakage](https://packetstormsecurity.com/files/172664/GS20230531163517.txt)
- [ ] [Qualcomm Adreno/KGSL Unchecked Cast…
-
### 🐛 Describe the bug
Hi! I tried to run the ppo_sentiments_llama example but got the error below.
```
╭─────────────────────────────── Traceback (most recent call last) ─────────────────────────…
-
### Required prerequisites
- [X] I have read the documentation .
- [X] I have searched the [Issue Tracker](https://github.com/PKU-Alignment/safe-rlhf/issues) and [Discussions](https://github.com/P…
-
Hi,
I tried to reproduce the whole process on a 8xV100 server with following command:
```bash
python train.py --actor-model facebook/opt-13b --reward-model facebook/opt-350m --num-gpus 8
```
Af…