search

BlinkDL / RWKV-LM

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
Apache License 2.0
12.32k stars 838 forks source link

txt数据集格式 #128

Closed cy565025164 closed 1 year ago

cy565025164 commented 1 year ago

“tokenize .txt into train.npy data”，.txt数据能给一些例子吗？

Leoeeeeeeea commented 1 year ago

i need too!!!

BlinkDL commented 1 year ago

语言模型是学习txt的内容。你想怎样使用模型，就怎样设计txt。

可以加【微调和训练QQ群 439087067】学习交流

BlinkDL commented 1 year ago

@cy565025164 @Leoeeeeeeea 请阅读 https://zhuanlan.zhihu.com/p/638326262 https://zhuanlan.zhihu.com/p/639629050