Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / horseee/LLM-Pruner issues and pull requests

#78 - 关于consecutive_groups

Issue - State: open - Opened by VincentZ-2020 7 days ago

#77 - UnicodeDecodeError: 'utf-8' codec can't decode byte 0x8b in position 1

Issue - State: closed - Opened by BrownTan 17 days ago - 4 comments

#76 - Taylor pruner under-utilizing resources

Issue - State: open - Opened by marianbasti 18 days ago

#74 - No such file or directory: pytorch_model.bin

Issue - State: open - Opened by yaolu-zjut 22 days ago - 2 comments

#73 - Can the pruned model be converted into a gguf file？

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

#72 - Custom Model pruning

Issue - State: open - Opened by saidineshpola 23 days ago

#71 - Does it support qwen2?

Issue - State: open - Opened by yangxue-1 24 days ago - 1 comment

#70 - Update ppl_dataset.py with trust_remote_code

Pull Request - State: open - Opened by manickavela29 26 days ago - 1 comment

#69 - Llama3 reports shape error after pruning

Issue - State: open - Opened by WentaoTan 28 days ago - 4 comments

#68 - Loading pruned model for causal llm

Issue - State: open - Opened by sriyachakravarthy about 1 month ago

#67 - Update hf_prune.py

Pull Request - State: open - Opened by aritralegndery about 1 month ago - 1 comment

#66 - I would like to ask if the current version is suitable for qwen.

Issue - State: open - Opened by wangxiaoxue about 2 months ago - 3 comments

#65 - Support GQA &Llama 3 & Llama 3.1.

Pull Request - State: closed - Opened by horseee about 2 months ago

#64 - Adaptation of GQA

Issue - State: open - Opened by junzhang-zj about 2 months ago - 7 comments

#62 - 请问能裁剪普通的transformer模型吗

Issue - State: open - Opened by SKY072410 3 months ago

#61 - 请问可以支持chatglm3剪枝吗

Issue - State: open - Opened by Franklin-L 3 months ago

#60 - Difference in Perplexity Values

Issue - State: open - Opened by nikhil-ghosh-berkeley 4 months ago

#59 - No random seed Settings found in post_training.py

Issue - State: open - Opened by JunKong5 4 months ago

#57 - Pruning llama3

Issue - State: open - Opened by yinwangsong 5 months ago

#56 - I tired Mistral 7b model, but I got this issue

Issue - State: open - Opened by TejasLidhure 5 months ago

#55 - How to prune the embedding and lm_head?

Issue - State: open - Opened by L-hongbin 6 months ago

#54 - Is this method implementable on multi-GPUs?

Issue - State: open - Opened by LeonCheng0129 6 months ago

#53 - RecursionError: maximum recursion depth exceeded

Issue - State: open - Opened by Zhenyu001225 6 months ago - 2 comments

#51 - 剪枝率值的问题

Issue - State: closed - Opened by qxpBlog 8 months ago

#50 - 延迟评估

Issue - State: closed - Opened by qxpBlog 8 months ago - 4 comments

#49 - The quantization of the compressed models

Issue - State: open - Opened by lihuang258 8 months ago

#47 - OSError: Can't load tokenizer for 'baffo32/decapoda-research-llama-7B-hf'.

Issue - State: closed - Opened by qxpBlog 8 months ago - 2 comments

#46 - Cannot use huggface to load

Issue - State: open - Opened by coderchem 9 months ago

#44 - cannot import name 'SiLUActivation' from 'transformers.activations'

Issue - State: closed - Opened by azuryl 9 months ago - 1 comment

#42 - Adding a tutorial for adapting new models?

Issue - State: open - Opened by zhichaoxu-shufe 10 months ago

#40 - Pruning MQA?

Issue - State: open - Opened by jianyuheng 10 months ago

#39 - Question related to the model tuning

Issue - State: open - Opened by shawnricecake 10 months ago - 2 comments

#38 - 为什么num_examples默认是10？

Issue - State: open - Opened by coderchem 10 months ago - 2 comments

#37 - the new pytorch.bin is bigger than original model issue

Issue - State: open - Opened by lb553024300 10 months ago - 4 comments

#35 - a post-training issue

Issue - State: open - Opened by cmnfriend 10 months ago - 2 comments

#34 - Reproducing paper results

Issue - State: open - Opened by grigorn 10 months ago - 6 comments

#33 - Latency code

Issue - State: closed - Opened by tuidan 10 months ago - 2 comments

#32 - Can not import LlamaConfig

Issue - State: closed - Opened by Ahmed-Roushdy 11 months ago - 1 comment

#31 - Examples on the Huggingface Hub

Issue - State: open - Opened by vgoklani 11 months ago

#30 - When will you support ChatGLM?

Issue - State: open - Opened by AboveParadise 11 months ago

#29 - Force even pruning across layers

Issue - State: open - Opened by thedarkzeno 12 months ago - 1 comment

#28 - Calculating Importance of 'param_mix'

Issue - State: closed - Opened by kiucho 12 months ago - 2 comments

#27 - When would the code for GPT-J-6B be released?

Issue - State: closed - Opened by mumuyeye 12 months ago - 1 comment

#26 - Added ability to load custom dataset

Pull Request - State: closed - Opened by michaelvin1322 about 1 year ago

#25 - Evaluation metric (acc vs. acc_norm) for lm-evaluation-harness tasks

Issue - State: closed - Opened by bokyeong1015 about 1 year ago - 4 comments

#24 - Sparse Mask question

Issue - State: closed - Opened by coldplayers about 1 year ago - 2 comments

#23 - param_first and param_mix result the same ppl

Issue - State: open - Opened by Kausal-Lei about 1 year ago - 1 comment

#22 - Error occurs when pruning LLaMa2-7b

Issue - State: closed - Opened by moonlightian about 1 year ago - 6 comments

#20 - Adding quantization

Issue - State: open - Opened by Duncan1115 about 1 year ago - 9 comments

#19 - evaluate

Issue - State: open - Opened by StevensPrime about 1 year ago - 1 comment

#18 - Pruning Llama2-7B

Issue - State: open - Opened by acalatrava about 1 year ago - 4 comments

#17 - recover training

Issue - State: open - Opened by xcg940123 about 1 year ago - 1 comment

#16 - Can prune model convert to llama.cpp ggml？

Issue - State: open - Opened by shaonianyr about 1 year ago - 1 comment

#15 - Checking the pruned but uncompressed model

Issue - State: open - Opened by ZN1010 about 1 year ago - 10 comments

#14 - Code for evaluation on zero-shot tasks using lm-evaluation-harness

Issue - State: closed - Opened by jameslahm about 1 year ago - 2 comments

#13 - Eval Loss NaN on Llama-2

Issue - State: open - Opened by mmichaelzhang about 1 year ago - 3 comments
Labels: question

#11 - Use LLM-Pruner for Baichuan model

Issue - State: closed - Opened by Daisy5296 about 1 year ago - 15 comments
Labels: bug

#10 - Zero-shot Evaluation

Issue - State: closed - Opened by Duncan1115 about 1 year ago - 2 comments

#9 - Gain using more data

Issue - State: closed - Opened by Daisy5296 about 1 year ago - 5 comments

#8 - When would the code for ChatGLM be released?

Issue - State: open - Opened by moonlightian about 1 year ago - 1 comment

#7 - RuntimeError of test_speedup.py

Issue - State: closed - Opened by Daisy5296 about 1 year ago - 9 comments

#6 - my process have some problems

Issue - State: closed - Opened by 18140663659 about 1 year ago - 1 comment

#5 - this method can be used for bloom?

Issue - State: open - Opened by 18140663659 about 1 year ago - 5 comments
Labels: enhancement

#4 - Error when using GPU for pruning

Issue - State: closed - Opened by Daisy5296 about 1 year ago - 3 comments

#3 - Reload the prunered Model failed

Issue - State: closed - Opened by Zhuqln over 1 year ago - 8 comments

#2 - Update README.md

Pull Request - State: closed - Opened by eltociear over 1 year ago

#1 - Question on recovery and training data

Issue - State: closed - Opened by Qubitium over 1 year ago - 2 comments