---
language:
- en
tags:
- pytorch
- text-generation
- causal-lm
- rwkv
license: apache-2.0
datasets:
- The Pile

---

# RWKV-4 3B

## Model Description

RWKV-4 3B is a L32-D2560 causal language model trained on the Pile. See https://github.com/BlinkDL/RWKV-LM for details.

** Note: It's a BF16 model, and it may overflow if you are using FP16 (probably fixable by rescaling the weights). **

At this moment you have to use my Github code (https://github.com/BlinkDL/RWKV-LM) to run it.

ctx_len = 1024
n_layer = 32
n_embd = 2560

Preview checkpoint: RWKV-4-Pile-3B-20220923-3822.pth : Trained on the Pile for 157B tokens.
* Pile loss 1.9857
* LAMBADA ppl 5.61, acc 61.87%
* PIQA acc 73.78%
* SC2016 acc 69.80%
* Hellaswag acc_norm 57.46%