Ecosyste.ms: Issues

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

GitHub / microsoft/DeepSpeedExamples issues and pull requests

#532 - Hyper-param tuning for PPO

Issue - State: open - Opened by luzai about 1 year ago

#529 - Much more memory used in step 3 when using multi gpus compared to using single gpu

Issue - State: open - Opened by cokuehuang about 1 year ago
Labels: deespeed chat, system

#528 - Rewards in ppo seem to be recomputed many times

Issue - State: open - Opened by dwyzzy about 1 year ago
Labels: deespeed chat, modeling

#527 - Step2 reward model 'chosen_last_scores' is really low, while acc is really high

Issue - State: open - Opened by LuciusMos about 1 year ago - 1 comment
Labels: question, deespeed chat, modeling

#526 - Step 3 issue - TypeError for * : float and NoneType for DeepSpeedChat while generating experience.

Issue - State: open - Opened by Ankush2k about 1 year ago - 1 comment
Labels: deespeed chat, new-config, modeling

#525 - [bug]AttributeError: 'DeepSpeedHybridEngine' object has no attribute 'mp_group'

Issue - State: open - Opened by qingchu123 about 1 year ago - 1 comment
Labels: bug, deespeed chat

#524 - RuntimeError: Error building extension 'transformer_inference'

Issue - State: closed - Opened by li995495592 about 1 year ago - 3 comments
Labels: deespeed chat

#523 - Lm workaround

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

#523 - Lm workaround

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

#522 - Training gpt-neo-1.3B on GPU

Issue - State: open - Opened by GenVr about 1 year ago - 2 comments
Labels: deespeed chat, system

#521 - OOM problem when fine-tune reward model with LLaMA in step 2

Issue - State: open - Opened by kiseliu about 1 year ago - 1 comment
Labels: deespeed chat, llama

#519 - fix spelling mistakes & bash script

Pull Request - State: open - Opened by QixuanAI about 1 year ago - 2 comments

#518 - RLHF model return '{: {: {:' of every input

Issue - State: open - Opened by kuangdao about 1 year ago - 1 comment
Labels: deespeed chat, modeling

#517 - Sequence truncation mistake in step3 training

Issue - State: open - Opened by puyuanOT about 1 year ago
Labels: deespeed chat, modeling

#517 - Sequence truncation mistake in step3 training

Issue - State: open - Opened by puyuanOT about 1 year ago
Labels: deespeed chat, modeling

#516 - no GPU resources available

Issue - State: closed - Opened by wuchaooooo about 1 year ago - 1 comment
Labels: deespeed chat

#516 - no GPU resources available

Issue - State: closed - Opened by wuchaooooo about 1 year ago - 1 comment
Labels: deespeed chat

#515 - RuntimeError: Error building extension 'transformer_inference'

Issue - State: closed - Opened by wwh5441 about 1 year ago - 1 comment

#514 - Does the framework support ChatGLM now?

Issue - State: open - Opened by MAJIN123 about 1 year ago - 2 comments
Labels: deespeed chat, modeling

#513 - ValidationError: 1 validation error for DeepSpeedZeroConfig

Issue - State: closed - Opened by Chtholly1 about 1 year ago - 5 comments

#513 - ValidationError: 1 validation error for DeepSpeedZeroConfig

Issue - State: closed - Opened by Chtholly1 about 1 year ago - 5 comments

#512 - In Step3, RuntimeError:RewardModel:size mismatch for rwtranrsformer.decoder.embed_tokens.weight

Issue - State: open - Opened by KyrieXu11 about 1 year ago - 6 comments
Labels: deespeed chat, llama

#512 - In Step3, RuntimeError:RewardModel:size mismatch for rwtranrsformer.decoder.embed_tokens.weight

Issue - State: open - Opened by KyrieXu11 about 1 year ago - 6 comments
Labels: deespeed chat, llama

#511 - DeepSpeed-Chat cannot load models from local file?

Issue - State: open - Opened by MianWang123 about 1 year ago
Labels: deespeed chat, new-config

#511 - DeepSpeed-Chat cannot load models from local file?

Issue - State: open - Opened by MianWang123 about 1 year ago
Labels: deespeed chat, new-config

#510 - How to save the model after each epoch

Issue - State: open - Opened by nieallen about 1 year ago - 1 comment
Labels: deespeed chat

#510 - How to save the model after each epoch

Issue - State: open - Opened by nieallen about 1 year ago - 1 comment
Labels: deespeed chat

#509 - NotImplementedError: Cannot copy out of meta tensor; no data!

Issue - State: closed - Opened by yangzhipeng1108 about 1 year ago - 1 comment

#509 - NotImplementedError: Cannot copy out of meta tensor; no data!

Issue - State: closed - Opened by yangzhipeng1108 about 1 year ago - 1 comment

#508 - Is there any Deepspeed Inference PTQ Example?

Issue - State: open - Opened by tingshua-yts about 1 year ago
Labels: question, deespeed chat

#508 - Is there any Deepspeed Inference PTQ Example?

Issue - State: open - Opened by tingshua-yts about 1 year ago
Labels: question, deespeed chat

#507 - Error for run_chinese.sh of step1, other_language

Issue - State: closed - Opened by korlin0110 about 1 year ago - 3 comments
Labels: deespeed chat

#507 - Error for run_chinese.sh of step1, other_language

Issue - State: closed - Opened by korlin0110 about 1 year ago - 3 comments
Labels: deespeed chat

#506 - In Step3: RuntimeError: numel: integer multiplication overflow

Issue - State: open - Opened by 480284856 about 1 year ago
Labels: deespeed chat, new-config, modeling

#506 - In Step3: RuntimeError: numel: integer multiplication overflow

Issue - State: open - Opened by 480284856 about 1 year ago
Labels: deespeed chat, new-config, modeling

#505 - Eextension of the issus #479, chatbot.py cannot load the bloom model

Issue - State: open - Opened by korlin0110 about 1 year ago - 2 comments
Labels: deespeed chat, new-config

#505 - Eextension of the issus #479, chatbot.py cannot load the bloom model

Issue - State: open - Opened by korlin0110 about 1 year ago - 2 comments
Labels: deespeed chat, new-config

#504 - merge master

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

#503 - Might be a bug of hibrid engine : In Step3 wrong generation secquence when hibrid engine is enabled.

Issue - State: open - Opened by laoda513 about 1 year ago - 4 comments
Labels: deespeed chat

#502 - Does the 1.3B model support multiple rounds of dialogue?

Issue - State: open - Opened by tensorflowt about 1 year ago
Labels: deespeed chat

#501 - Fixing the numerical instability when calculating the loss of the cri…

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

#498 - CPU OOM in training of step3

Issue - State: open - Opened by cokuehuang about 1 year ago - 2 comments
Labels: deespeed chat, system

#498 - CPU OOM in training of step3

Issue - State: open - Opened by cokuehuang about 1 year ago - 2 comments
Labels: deespeed chat, system

#497 - deepspeed hybrid-engine support bloom model with zero3?

Issue - State: open - Opened by null-test-7 about 1 year ago - 1 comment
Labels: deespeed chat, new-config

#496 - What is the purpose of end_of_conversation_token="<|endoftext|> and why it is not added as special token?

Issue - State: open - Opened by REIGN12 about 1 year ago - 1 comment
Labels: question, deespeed chat

#495 - What's default data_path in single_gpu/run_1.3b.sh

Issue - State: closed - Opened by zy-sunshine about 1 year ago - 1 comment
Labels: question, deespeed chat

#493 - Does deepspeed-chat support LLama ？

Issue - State: open - Opened by janelu9 about 1 year ago
Labels: question, deespeed chat, llama

#493 - Does deepspeed-chat support LLama ？

Issue - State: open - Opened by janelu9 about 1 year ago
Labels: question, deespeed chat, llama

#492 - OPT-13b model not pass step3_rlhf_finetuning

Issue - State: open - Opened by korlin0110 about 1 year ago - 3 comments
Labels: bug, deespeed chat

#491 - need a better metrics other than acc and average score for the reward modeling step

Issue - State: open - Opened by DanqingZ about 1 year ago - 5 comments
Labels: enhancement, deespeed chat, modeling

#491 - need a better metrics other than acc and average score for the reward modeling step

Issue - State: open - Opened by DanqingZ about 1 year ago - 5 comments
Labels: enhancement, deespeed chat, modeling

#490 - Why is it a need to set the pad token to be equal to eos token?

Issue - State: open - Opened by alibabadoufu about 1 year ago - 1 comment
Labels: question, deespeed chat, modeling

#490 - Why is it a need to set the pad token to be equal to eos token?

Issue - State: open - Opened by alibabadoufu about 1 year ago - 1 comment
Labels: question, deespeed chat, modeling

#489 - The reward in step3 seems to be completely random without any noticeable increase.

Issue - State: open - Opened by laoda513 about 1 year ago - 7 comments
Labels: deespeed chat, modeling

#488 - Any plans to upgrade triton version in stable-diffusion example?

Issue - State: open - Opened by alexeigor about 1 year ago
Labels: question

#488 - Any plans to upgrade triton version in stable-diffusion example?

Issue - State: open - Opened by alexeigor about 1 year ago
Labels: question

#487 - program stop when i run my run_chinese.sh in GPU A100*4 80G

Issue - State: open - Opened by liyuyuan6969 about 1 year ago - 3 comments
Labels: deespeed chat

#487 - program stop when i run my run_chinese.sh in GPU A100*4 80G

Issue - State: open - Opened by liyuyuan6969 about 1 year ago - 3 comments
Labels: deespeed chat

#483 - Error in step1 finetuning decapoda-research/llama-7b-hf

Issue - State: open - Opened by yctam about 1 year ago - 3 comments
Labels: deespeed chat

#482 - A100 40 GB: OOM on step-3 for opt-6.7B

Issue - State: open - Opened by akashsaravanan-georgian about 1 year ago - 4 comments
Labels: deespeed chat, system, new-config

#481 - RuntimeError: Error building extension 'transformer_inference' in step3

Issue - State: closed - Opened by SunQiDong1999 about 1 year ago - 3 comments

#480 - unable to load 4 7b size model in step3

Issue - State: open - Opened by Mr-lonely0 about 1 year ago - 3 comments
Labels: deespeed chat, system, new-config

#479 - Can not use bloom-560m model in the step2_reward_model_finetuning

Issue - State: open - Opened by korlin0110 about 1 year ago - 3 comments
Labels: deespeed chat, system, new-config

#479 - Can not use bloom-560m model in the step2_reward_model_finetuning

Issue - State: open - Opened by korlin0110 about 1 year ago - 3 comments
Labels: deespeed chat, system, new-config

#478 - inference OOM

Issue - State: open - Opened by Haoran1234567 about 1 year ago - 1 comment
Labels: deespeed chat

#478 - inference OOM

Issue - State: open - Opened by Haoran1234567 about 1 year ago - 1 comment
Labels: deespeed chat

#475 - DeepspeedExample-Chat training failed with single_node parameter

Issue - State: closed - Opened by wanbo432503 about 1 year ago - 11 comments

#474 - PPO training unable to reproduce the training log provided

Issue - State: open - Opened by REIGN12 about 1 year ago
Labels: deespeed chat, modeling

#474 - PPO training unable to reproduce the training log provided

Issue - State: open - Opened by REIGN12 about 1 year ago
Labels: deespeed chat, modeling

#472 - [deepspeed-chat] finetuned model can not even overfit a really small dataset with only 11 samples

Issue - State: open - Opened by valkryhx about 1 year ago - 1 comment
Labels: deespeed chat, modeling

#472 - [deepspeed-chat] finetuned model can not even overfit a really small dataset with only 11 samples

Issue - State: open - Opened by valkryhx about 1 year ago - 1 comment
Labels: deespeed chat, modeling

#470 - Fix to allow hf-generation to generate eos-token

Pull Request - State: open - Opened by aakutalev about 1 year ago - 4 comments

#470 - Fix to allow hf-generation to generate eos-token

Pull Request - State: open - Opened by aakutalev about 1 year ago - 4 comments

#469 - Training time of DeepSpeed-Chat’s RLHF examples

Issue - State: open - Opened by zasitonbyl about 1 year ago - 7 comments
Labels: deespeed chat, modeling

#469 - Training time of DeepSpeed-Chat’s RLHF examples

Issue - State: open - Opened by zasitonbyl about 1 year ago - 7 comments
Labels: deespeed chat, modeling

#467 - KeyError: "_name_or_path" in locally loading tokenizer config file for DeepSpeed-Chat

Issue - State: open - Opened by ccclyu about 1 year ago - 1 comment
Labels: deespeed chat

#463 - May I run any one of the example of deepseed on one 1080ti, P40 and P100

Issue - State: closed - Opened by SeekPoint about 1 year ago - 3 comments
Labels: deespeed chat

#463 - May I run any one of the example of deepseed on one 1080ti, P40 and P100

Issue - State: closed - Opened by SeekPoint about 1 year ago - 3 comments
Labels: deespeed chat

#458 - Adding two loss from actor will lead to an error " gradient computed twice for this partition"

Issue - State: open - Opened by piekey1994 about 1 year ago - 1 comment
Labels: deespeed chat

#458 - Adding two loss from actor will lead to an error " gradient computed twice for this partition"

Issue - State: open - Opened by piekey1994 about 1 year ago - 1 comment
Labels: deespeed chat

#456 - enable_hybrid_engine issue

Issue - State: open - Opened by llllooong about 1 year ago - 3 comments
Labels: deespeed chat

#456 - enable_hybrid_engine issue

Issue - State: open - Opened by llllooong about 1 year ago - 3 comments
Labels: deespeed chat

#454 - fix step_time

Pull Request - State: open - Opened by thuzhf about 1 year ago

#454 - fix step_time

Pull Request - State: open - Opened by thuzhf about 1 year ago

#453 - [BUG]RuntimeError: CUDA error: unknown error

Issue - State: open - Opened by SH0AN about 1 year ago - 3 comments
Labels: deespeed chat

#453 - [BUG]RuntimeError: CUDA error: unknown error

Issue - State: open - Opened by SH0AN about 1 year ago - 3 comments
Labels: deespeed chat

#452 - cant use zero-offload

Issue - State: open - Opened by yanqiangmiffy about 1 year ago - 4 comments
Labels: deespeed chat

#452 - cant use zero-offload

Issue - State: open - Opened by yanqiangmiffy about 1 year ago - 4 comments
Labels: deespeed chat

#451 - Finetuning Bloom model in step 3 failed

Issue - State: open - Opened by cokuehuang about 1 year ago - 5 comments
Labels: deespeed chat

#447 - training 12b model seems to require more memory than expected

Issue - State: open - Opened by ChaoChungWu-Johnson about 1 year ago - 2 comments
Labels: deespeed chat, new-config

#447 - training 12b model seems to require more memory than expected

Issue - State: open - Opened by ChaoChungWu-Johnson about 1 year ago - 2 comments
Labels: deespeed chat, new-config

#443 - How to train deepspeed-chat using nccl with multi-nodes?

Issue - State: open - Opened by SefaZeng about 1 year ago - 2 comments
Labels: deespeed chat

#443 - How to train deepspeed-chat using nccl with multi-nodes?

Issue - State: open - Opened by SefaZeng about 1 year ago - 2 comments
Labels: deespeed chat

#442 - Performance gap between actor ema and actor

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