ridgerchu/matmulfreellm issues and pull requests

#42 - AttributeError: module 'triton.language.math' has no attribute 'round'

Issue - State: open - Opened by sourcesync 4 days ago - 2 comments

#41 - Added Example Notebook and Working Docker File

Pull Request - State: closed - Opened by jnesfield 6 days ago - 1 comment

#40 - Question about ”fused_recurrent_hgrn“ function

Issue - State: closed - Opened by ianwangwei about 1 month ago - 2 comments

#39 - training model with huggingface

Issue - State: open - Opened by KETTY2 about 1 month ago - 2 comments

#38 - triton error using generate.py

Issue - State: closed - Opened by KETTY2 about 1 month ago - 2 comments

#37 - Discrepency in code and paper related to HGRNBitAttention

Issue - State: open - Opened by loki-r about 2 months ago - 1 comment

#36 - About the FPGA Performance Metrics

Issue - State: open - Opened by SinnLiu about 2 months ago - 2 comments

#35 - Can Meta's llama model be quantized to matmulfree?

Issue - State: closed - Opened by impredicative about 2 months ago - 1 comment

#34 - LLVM ERROR: Cannot select: intrinsic %llvm.nvvm.shfl.sync.bfly.i32

Issue - State: open - Opened by jeisonmp 3 months ago - 5 comments

#33 - Larger models available or planned?

Issue - State: open - Opened by jaggzh 3 months ago - 5 comments

#32 - Any pretrained model bigger than 2.7B?

Issue - State: closed - Opened by impredicative 3 months ago - 3 comments

#31 - How do I train my models?

Issue - State: open - Opened by zxk907 3 months ago - 8 comments

#30 - Reproduce the results of the code in the paper

Issue - State: open - Opened by hsb1995 3 months ago - 4 comments

#29 - docs: update README.md

Pull Request - State: closed - Opened by eltociear 3 months ago

#28 - Minor fixes

Pull Request - State: open - Opened by TeaPoly 3 months ago

#27 - Convert existing model to mmfree

Issue - State: open - Opened by DewEfresh 3 months ago - 1 comment

#26 - MLGRU

Issue - State: open - Opened by AACengineer 3 months ago - 1 comment

#25 - Update README.md

Pull Request - State: closed - Opened by jeshraghian 3 months ago

#24 - scaling_law.png

Issue - State: open - Opened by clort81 3 months ago - 1 comment

#23 - how to train or ft?

Issue - State: open - Opened by dongjicheng 3 months ago - 4 comments

#22 - Does matmulfreellm support Windows 10?

Issue - State: open - Opened by fangkuoyu 3 months ago - 6 comments

#21 - Why Transformer++

Issue - State: open - Opened by sdmorrey 3 months ago - 1 comment

#20 - Baseline weights?

Issue - State: closed - Opened by justanhduc 3 months ago - 2 comments

#19 - why FusedBitLinear.forward() use F.linear() with float16 inputs?

Issue - State: open - Opened by AACengineer 3 months ago - 3 comments

#18 - Ternary weight values

Issue - State: closed - Opened by dmahurin 3 months ago - 5 comments

#17 - No reduction in VRAM usage

Issue - State: open - Opened by radna0 3 months ago - 5 comments

#10 - FPGA implementation

Issue - State: open - Opened by ws200106 3 months ago - 4 comments

#9 - Failure to train.

Issue - State: closed - Opened by kazuya-hodatsu-336-1 3 months ago - 4 comments

#8 - Error on pip install: ModuleNotFoundError: No module named 'torch'

Issue - State: closed - Opened by josegomezr 3 months ago - 2 comments

#7 - Fix bitnet ops missing class definition

Pull Request - State: closed - Opened by awentzonline 3 months ago - 1 comment

Ecosyste.ms: Issues

GitHub / ridgerchu/matmulfreellm issues and pull requests