Issues
- 4
Full version of H3 model?
#29 opened by FeelingFatigued - 2
use_fast_fftconv generates error
#28 opened by FeelingFatigued - 2
ERROR: CUDA RT call cudaFuncSetAttribute. Failed with invalid device function (98).
#27 opened by wang-zerui - 1
Setup for repo unclear
#26 opened by inf3rnus - 2
- 0
- 2
why dividing kv_f by fft_size?
#22 opened by violet-zct - 2
- 5
Release of pretraining and fine tuning code
#20 opened by ksrinivs64 - 3
- 2
- 3
Error Running `generate_text_h3.py` (`CUDA error: CUBLAS_STATUS_NOT_INITIALIZED`)
#17 opened by gdebayan - 9
Having trouble for compiling fftconv
#18 opened by Doraemonzzz - 3
Releasing Training and Synthetic benchmarks
#12 opened by mojanjp - 4
Training code?
#6 opened by NtaylorOX - 4
- 2
2.7B Evaluations
#1 opened by sdtblck - 4
- 3
- 1
FFT Conv on Seq > 8192?
#14 opened by darius-lam - 2
What is fftconv_bwd doing?
#13 opened by darius-lam - 1
Unpickling errors when loading models
#11 opened by mttk - 4
Correct method to load 2.7B?
#10 opened by BlinkDL - 2
Trying to generate something coherent
#8 opened by nikitastaf1996 - 1
Licensing information
#5 opened by augmented-fog - 2
ssm utils
#2 opened by bryanhpchiang