Related Issues (20)
- [BUG] Deepspeed zero3 error in loading pretrained model by transformers.LlamaForCausalLM.from_pretrained function. HOT 2
- [BUG] FlopsProfiler.start_profile() hangs when using FSDP HOT 2
- [REQUEST] Make DeepSpeed FastGen support tokens as input?
- [BUG]Problems with model initialization when using PipelineModule
- [BUG] Deepspeed zero-3 giving an OOM error HOT 5
- ValueError: Only able to place 0 replicas, but 2 replicas were requested. HOT 2
- [BUG] Llama-7b + peft + Deepspeed + Azure ML command job (zero 2 and zero3) OOM error during deepspeed.initialize
- [BUG] hard coded to CUDA in builder.py HOT 4
- [REQUEST] Democratizing Present Day LLM Fine-Tuning To The GPU Poor HOT 2
- RuntimeError: Error building extension 'fused_adam'[BUG] HOT 10
- [BUG] CUDA_VISIBLE_DEVICES is not parsed correctly
- [BUG] Deepspeed.initialize: 64Gb VRAM arent enough to PEFT llama 7b?
- Accelerator Abstraction: Fix skipped tests:
- [REQUEST] is there a plan to support PagedAttention in deepspeed-inference.v2? Thanks
- [BUG] Order of partitioning in universal checkpoint HOT 12
- [REQUEST] Ada-Lovelace support HOT 2
- [BUG] MegatronMoELayerPolicy not mapped
- [BUG]Deepspeed OOM out of CPU memory on A100 80Gx4. The more GPU use the more CPU memory ds occupy. HOT 4
- **Update**: We have investigate further. The norm difference is caused by the gradient difference. Our current solution is to turn the overlap_comm to False when using Zero2. This fixes the issue, however, makes the training slower. @tjruwase hope this may help.
- [BUG]when i use deepspeed to finetune a model, the cuda memory oom because input a 128k data. How can i start training.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from deepspeed.