Downloads · 30 days
335
21% of all-time downloads
rwkv-3-pile-169m
BlinkDL/rwkv-3-pile-169m
rwkv-3-pile-169m is a text generation model from BlinkDL. Use it when you need the model to write or continue text. It is set up for rwkv. The card lists the license as apache-2.0.
RWKV-3 169M is a L12-D768 causal language model trained on the Pile. See https://github.com/BlinkDL/RWKV-LM for details.
All-time downloads
1.6K
Public
Repo size
1.4 GB
Likes
6
Public
Repo makeup
Click a slice to open those files.
.pth1.4 GB · 100%
At a glance
- Task
- Text Generation
- Library
- rwkv
- License
- apache-2.0
- Access
- Public
- Created
- Jul 6, 2022
- Updated
- Sep 14, 2022
- SHA
- f9a6ddeb
Try a prompt
Model card
From the Hugging Face model README
RWKV-3 169M
Model Description
RWKV-3 169M is a L12-D768 causal language model trained on the Pile. See https://github.com/BlinkDL/RWKV-LM for details.
At this moment you have to use my Github code (https://github.com/BlinkDL/RWKV-v2-RNN-Pile) to run it.
ctx_len = 768 n_layer = 12 n_embd = 768
Final checkpoint: RWKV-3-Pile-20220720-10704.pth : Trained on the Pile for 328B tokens.
- Pile loss 2.5596
- LAMBADA ppl 28.82, acc 32.33%
- PIQA acc 64.15%
- SC2016 acc 57.88%
- Hellaswag acc_norm 32.45%
Preview checkpoint: 20220703-1652.pth : Trained on the Pile for 50B tokens. Pile loss 2.6375, LAMBADA ppl 33.30, acc 31.24%.
- Task
- Text Generation
- Library
- rwkv
- License
- apache-2.0
- Languages
- en
- Created
- Jul 6, 2022
- Updated
- Sep 14, 2022