Arcman

The pre-trained model can't access

2

The connection refused I mean.

**Describe the bug** A clear and concise description of what the bug is and what the main root cause error is. Test very thoroughly before submitting. **To Reproduce** Steps to...

rwkv-4 RWKV_TimeMix如何分割序列，没有看到cuda_run接受任何sequence切分的标志？

1

RWKV_TimeMix中在序列维度上进行操作，在进行训练时训练数据常常是首尾相接的，序列之间需要隔断分开进行处理，例如flashattention会接收一个序列开始位置的输入，RUN_CUDA似乎没有，是如何实现的