kyegomez/bitnet issues and pull requests

#75 - [BUG] Is activation actually quantized? (I think it's still in f16 or f32)

Issue - State: open - Opened by burichh 15 days ago
Labels: bug

#74 - Hint for 1bit Multi-head attention

Issue - State: closed - Opened by CastleJin 18 days ago - 2 comments

#73 - Bitnet RWKV-7

Issue - State: open - Opened by snapo 22 days ago

#72 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.12.4

Pull Request - State: open - Opened by dependabot[bot] 24 days ago
Labels: dependencies, github_actions

#71 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.12.3

Pull Request - State: closed - Opened by dependabot[bot] about 1 month ago - 1 comment
Labels: dependencies, github_actions

#70 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.12.2

Pull Request - State: closed - Opened by dependabot[bot] 3 months ago - 2 comments
Labels: dependencies, github_actions, no-pr-activity

#69 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.11.0

Pull Request - State: closed - Opened by dependabot[bot] 4 months ago - 1 comment
Labels: dependencies, github_actions

#68 - I wonder what hardware conditions (GPU) the code uses, and why the loss value has been above 5.2 after running the train.py file, and the validation generated unreadable incomprehensible content.

Issue - State: closed - Opened by Dayun0925 4 months ago - 1 comment
Labels: bug, no-issue-activity

#67 - CUDA Optimization

Issue - State: closed - Opened by simulanics 4 months ago - 1 comment
Labels: no-issue-activity

#66 - Can it be used in Huggingface models such as stable diffusion and text to image or LLM models

Issue - State: closed - Opened by libai-lab 4 months ago - 1 comment
Labels: no-issue-activity

#65 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.10.3

Pull Request - State: closed - Opened by dependabot[bot] 5 months ago - 1 comment
Labels: dependencies, github_actions

#64 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.10.2

Pull Request - State: closed - Opened by dependabot[bot] 5 months ago - 1 comment
Labels: dependencies, github_actions

#63 - Question: embeddings 3bits?

Issue - State: closed - Opened by telamon 5 months ago - 1 comment
Labels: no-issue-activity

#62 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.10.1

Pull Request - State: closed - Opened by dependabot[bot] 5 months ago - 1 comment
Labels: dependencies, github_actions

#61 - Bump pypa/gh-action-pypi-publish from 1.9.0 to 1.10.0

Pull Request - State: closed - Opened by dependabot[bot] 6 months ago - 1 comment
Labels: dependencies, github_actions

#60 - [BUG] cant compile the cuda

Issue - State: closed - Opened by huynhducloi00 7 months ago - 2 comments
Labels: bug, no-issue-activity

#59 - Fix: Weight quantization sign should be the last operation

Pull Request - State: closed - Opened by jmbrito01 8 months ago - 1 comment
Labels: no-pr-activity

#58 - Create a javascript version that can run in a web browser?

Issue - State: closed - Opened by flatsiedatsie 8 months ago - 2 comments
Labels: no-issue-activity

#57 - NanoGPT sample

Issue - State: closed - Opened by izaxon 8 months ago - 1 comment
Labels: no-issue-activity

#56 - Bump pypa/gh-action-pypi-publish from 1.8.14 to 1.9.0

Pull Request - State: closed - Opened by dependabot[bot] 8 months ago
Labels: dependencies, github_actions

#56 - Bump pypa/gh-action-pypi-publish from 1.8.14 to 1.9.0

Pull Request - State: closed - Opened by dependabot[bot] 8 months ago
Labels: dependencies, github_actions

#55 - BitNet model performs wrose than Base Transformer

Issue - State: closed - Opened by johanssontan 9 months ago - 2 comments
Labels: bug, no-issue-activity

#54 - added if else statement to handle post_act_ln

Pull Request - State: closed - Opened by Hiromasa-H 9 months ago - 1 comment

#54 - added if else statement to handle post_act_ln

Pull Request - State: closed - Opened by Hiromasa-H 9 months ago - 1 comment

#53 - [BUG] BitFeedForward(post_act_ln=False) results in a TypeError

Issue - State: closed - Opened by Hiromasa-H 9 months ago - 3 comments
Labels: bug

#53 - [BUG] BitFeedForward(post_act_ln=False) results in a TypeError

Issue - State: closed - Opened by Hiromasa-H 9 months ago - 3 comments
Labels: bug

#52 - Expected BitLinear weight to be 1 or -1

Issue - State: closed - Opened by sanjeev-bhandari 10 months ago - 4 comments
Labels: no-issue-activity

#51 - what is the purpose of detach here?

Issue - State: closed - Opened by Weitian-Wang-Bosch 10 months ago - 2 comments

#51 - what is the purpose of detach here?

Issue - State: closed - Opened by Weitian-Wang-Bosch 10 months ago - 2 comments

#50 - Fix error in bitlinear algorithm

Pull Request - State: closed - Opened by Mrw33554432 10 months ago - 1 comment
Labels: no-pr-activity

#50 - Fix error in bitlinear algorithm

Pull Request - State: closed - Opened by Mrw33554432 10 months ago - 1 comment
Labels: no-pr-activity

#49 - [BUG] Bitnet Example Bug

Issue - State: closed - Opened by sneilan 11 months ago - 6 comments
Labels: bug, no-issue-activity

#49 - [BUG] Bitnet Example Bug

Issue - State: closed - Opened by sneilan 11 months ago - 6 comments
Labels: bug, no-issue-activity

#48 - Consider techniques from official training paper

Issue - State: closed - Opened by EwoutH 11 months ago - 2 comments
Labels: no-issue-activity

#48 - Consider techniques from official training paper

Issue - State: closed - Opened by EwoutH 11 months ago - 2 comments
Labels: no-issue-activity

#47 - fix grouping in bitlinear.py

Pull Request - State: closed - Opened by Jiangxg 11 months ago

#47 - fix grouping in bitlinear.py

Pull Request - State: closed - Opened by Jiangxg 11 months ago

#46 - 1.58bit algorithm implement recommend

Issue - State: closed - Opened by princepride 11 months ago - 2 comments
Labels: no-issue-activity

#46 - 1.58bit algorithm implement recommend

Issue - State: closed - Opened by princepride 11 months ago - 2 comments
Labels: no-issue-activity

#45 - Encountering Size Mismatch Error in Updated Code

Issue - State: closed - Opened by anonymousA123 11 months ago - 3 comments
Labels: bug, no-issue-activity

#45 - Encountering Size Mismatch Error in Updated Code

Issue - State: closed - Opened by anonymousA123 11 months ago - 3 comments
Labels: bug, no-issue-activity

#44 - Bump pypa/gh-action-pypi-publish from 1.8.12 to 1.8.14

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: dependencies, github_actions

#44 - Bump pypa/gh-action-pypi-publish from 1.8.12 to 1.8.14

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: dependencies, github_actions

#43 - [BUG] NoneType in sequential module in bit_ffn

Issue - State: closed - Opened by jayUyang 12 months ago - 1 comment
Labels: bug, no-issue-activity

#43 - [BUG] NoneType in sequential module in bit_ffn

Issue - State: closed - Opened by jayUyang 12 months ago - 1 comment
Labels: bug, no-issue-activity

#42 - [BUG] bitlinear fix

Issue - State: closed - Opened by jayUyang 12 months ago - 5 comments
Labels: bug, no-issue-activity

#42 - [BUG] bitlinear fix

Issue - State: closed - Opened by jayUyang 12 months ago - 5 comments
Labels: bug, no-issue-activity

#41 - is this actually working?

Issue - State: closed - Opened by fblgit 12 months ago - 11 comments
Labels: no-issue-activity

#41 - is this actually working?

Issue - State: closed - Opened by fblgit 12 months ago - 11 comments
Labels: no-issue-activity

#40 - Issue with model size after replacing BitLinear layer into a HF model (say Llama2-7b-chat)[BUG]

Issue - State: closed - Opened by mriganktiwari 12 months ago - 3 comments
Labels: bug, no-issue-activity

#39 - Revert "Jp"

Pull Request - State: closed - Opened by kyegomez 12 months ago

#39 - Revert "Jp"

Pull Request - State: closed - Opened by kyegomez 12 months ago

#38 - Google Drive Link to model weights is broken

Issue - State: closed - Opened by SinanAkkoyun 12 months ago - 1 comment
Labels: bug

#38 - Google Drive Link to model weights is broken

Issue - State: closed - Opened by SinanAkkoyun 12 months ago - 1 comment
Labels: bug

#37 - Fixed shape of beta and gamma for proper broadcasting

Pull Request - State: closed - Opened by dariocazzani 12 months ago

#37 - Fixed shape of beta and gamma for proper broadcasting

Pull Request - State: closed - Opened by dariocazzani 12 months ago

#36 - Requesting a Text-to-Text translation example

Issue - State: closed - Opened by TMammadov 12 months ago - 1 comment
Labels: no-issue-activity

#35 - The output of BitLinear is quite abnormal

Issue - State: closed - Opened by Jiangxg 12 months ago - 6 comments
Labels: bug

#34 - ImportError: cannot import name 'BitLinear15bs' from 'bitnet.bitbnet_b158'[BUG]

Issue - State: closed - Opened by Bobby-youngking 12 months ago - 6 comments
Labels: bug

#34 - ImportError: cannot import name 'BitLinear15bs' from 'bitnet.bitbnet_b158'[BUG]

Issue - State: closed - Opened by Bobby-youngking 12 months ago - 6 comments
Labels: bug

#33 - Update bitlinear.py

Pull Request - State: closed - Opened by ramonpeter 12 months ago - 3 comments

#33 - Update bitlinear.py

Pull Request - State: closed - Opened by ramonpeter 12 months ago - 3 comments

#32 - [BUG] residual connection wrong?

Issue - State: closed - Opened by qianlong0502 12 months ago - 1 comment
Labels: bug

#32 - [BUG] residual connection wrong?

Issue - State: closed - Opened by qianlong0502 12 months ago - 1 comment
Labels: bug

#31 - Bump pypa/gh-action-pypi-publish from 1.8.11 to 1.8.12

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: dependencies, github_actions

#31 - Bump pypa/gh-action-pypi-publish from 1.8.11 to 1.8.12

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: dependencies, github_actions

#30 - Update bitlinear.py

Pull Request - State: closed - Opened by ramonpeter 12 months ago - 2 comments

#30 - Update bitlinear.py

Pull Request - State: closed - Opened by ramonpeter 12 months ago - 2 comments

#29 - Jp

Pull Request - State: closed - Opened by Sunwood-ai-labs 12 months ago - 1 comment

#29 - Jp

Pull Request - State: closed - Opened by Sunwood-ai-labs 12 months ago - 1 comment

#28 - fix inference bug

Pull Request - State: closed - Opened by shi3z 12 months ago - 1 comment
Labels: no-pr-activity

#28 - fix inference bug

Pull Request - State: closed - Opened by shi3z 12 months ago - 1 comment
Labels: no-pr-activity

#27 - does not have support for mistral, gemma, etc and generate error [BUG] ?

Issue - State: closed - Opened by NickyDark1 12 months ago - 5 comments
Labels: bug, no-issue-activity

#27 - does not have support for mistral, gemma, etc and generate error [BUG] ?

Issue - State: closed - Opened by NickyDark1 12 months ago - 5 comments
Labels: bug, no-issue-activity

#26 - Parts of the BitLinear code doesn't match paper (before bit1.58)

Issue - State: closed - Opened by qqqllppp 12 months ago - 2 comments
Labels: no-issue-activity

#26 - Parts of the BitLinear code doesn't match paper (before bit1.58)

Issue - State: closed - Opened by qqqllppp 12 months ago - 2 comments
Labels: no-issue-activity

#25 - Question about weight quantization methodology memory savings

Issue - State: closed - Opened by nnethercott 12 months ago - 1 comment
Labels: no-issue-activity

#25 - Question about weight quantization methodology memory savings

Issue - State: closed - Opened by nnethercott 12 months ago - 1 comment
Labels: no-issue-activity

#24 - [BUG]multi-head attention is noop for BITLINEAR

Issue - State: closed - Opened by Bsdnbo 12 months ago - 1 comment
Labels: bug

#24 - [BUG]multi-head attention is noop for BITLINEAR

Issue - State: closed - Opened by Bsdnbo 12 months ago - 1 comment
Labels: bug

#23 - [BUG] Loss drops, model still produces gibberish?

Issue - State: closed - Opened by MichelNivard 12 months ago - 6 comments
Labels: bug, no-issue-activity

#22 - where to download bitnet model ?

Issue - State: closed - Opened by dibu28 12 months ago - 4 comments
Labels: bug

#22 - where to download bitnet model ?

Issue - State: closed - Opened by dibu28 12 months ago - 4 comments
Labels: bug

#21 - About 'replace_hf.py'

Issue - State: closed - Opened by chyoob 12 months ago - 3 comments
Labels: no-issue-activity

#21 - About 'replace_hf.py'

Issue - State: closed - Opened by chyoob 12 months ago - 3 comments
Labels: no-issue-activity

#20 - [BUG] Google drive link in readme.md is dead

Issue - State: closed - Opened by nathanielhudson 12 months ago - 1 comment
Labels: bug

#20 - [BUG] Google drive link in readme.md is dead

Issue - State: closed - Opened by nathanielhudson 12 months ago - 1 comment
Labels: bug

#19 - [BUG] Tensor size mismatch from train.py

Issue - State: closed - Opened by richardburleigh 12 months ago - 7 comments
Labels: bug

#19 - [BUG] Tensor size mismatch from train.py

Issue - State: closed - Opened by richardburleigh 12 months ago - 7 comments
Labels: bug

#18 - [BUG] Can't install with pipenv, pip

Issue - State: closed - Opened by calliope-pro about 1 year ago - 1 comment
Labels: bug

#18 - [BUG] Can't install with pipenv, pip

Issue - State: closed - Opened by calliope-pro about 1 year ago - 1 comment
Labels: bug

#17 - need a distributed training example

Issue - State: closed - Opened by sosofun about 1 year ago - 3 comments

#17 - need a distributed training example

Issue - State: closed - Opened by sosofun about 1 year ago - 3 comments

#16 - [BUG] I tried using BitLinear in nanoRWKV but got the error.

Issue - State: closed - Opened by win10ogod about 1 year ago - 3 comments
Labels: bug