RWKV-LM · Issues· 158 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #351
Verify evals on Papers with Code
Updated Sep 14, 2026 - #338
Can RWKV state encode abstract behavioral dispositions? | RWKV state 能否编码抽象行为倾向？
Updated Aug 28, 2026 - #349
Structured experiments on G1h 2.9B: L_state trajectory reg, N/K/mode effort frontier, causal state probe, multi-slot capacity
Updated Aug 14, 2026 - #343
RWKV-v7/rwkv_v7_numpy.py fails on NumPy <2.0 (`.mT` attribute)
Updated Aug 1, 2026 - #347
67
Updated Jul 28, 2026 - #346
Out-of-range token IDs trigger a CUDA device-side assert instead of a validation error
Updated Jul 26, 2026 - #339
Рэпчина
Updated Jun 5, 2026 - #337
Add support for true unlimited context
Updated May 16, 2026 - #332
RWKV_JIT_ON=0` is ignored when `RWKV_V7_ON=1
Updated Apr 28, 2026 - #335
Solution to the gating issue in RNN models.
Updated Apr 25, 2026 - #330
token填充问题
Updated Jan 30, 2026 - #329
Is this website, `www.rwkv.cn`, yours?
Updated Jan 22, 2026 - #326
A cleaner Read Me file
Updated Nov 11, 2025 - #317
log-linear-attention colab?
Updated Nov 11, 2025 - #323
Add support for pytorch for both training and interference
Updated Nov 11, 2025