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

169M模型在下游任务微调时效果不佳 #124

Closed zhanghanweii closed 1 year ago

zhanghanweii commented 1 year ago

我是用10w条数据进行下游任务微调时(恢复标点和大小写)，+qa等各种方法生成的回答都是随机的

BlinkDL commented 1 year ago

你需要使用正确的 tokenizer。请加Q群 439087067 学习