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
jsonl2binidx tool for new RWKV tokenizer (world vocab)? #134
I'm planning future training based on RWKV-v4-World series. Is there any tool for converting jsonl to binidx using RWKV World vocab-tokenizer?