Comments (3)
LvanderGoten
commented on May 9, 2024
I second that
from deepspeed.
jeffra
commented on May 9, 2024
We're working on fixing this soon. However, in my case I was able to get around this by setting a smaller initial scale power. See the 'initial_scale_power' fp16 parameter here: https://github.com/microsoft/DeepSpeed/blob/master/docs/config_json.md#fp16-training-options
from deepspeed.
ShadenSmith
commented on May 9, 2024
@jeffra I believe PR #166 addresses this issue. Let me know what you think.
from deepspeed.
Related Issues (20)
- [BUG] No `universal_checkpoint_info` in the Accelerate+Deepspeed Checkpoint HOT 6
- nv-nightly CI test failure HOT 1
- [BUG] (flops_profiler) Duplicate registration check for start_time_hook is not working
- [BUG: Whisper model pipeline parallel training] logits and ground truth size mismatch during loss calculation
- [Q&A] Why Deepspeed Ulysses could support long sequence length?
- Why not save frozen params unless: `self.zero_optimization_stage() >= ZeroStageEnum.gradients`? HOT 2
- [REQUEST]I do not understand the meaning of ' reduction ' in the ZERO++ paper.
- Deepspeed module not being able to install in the WSL environment HOT 2
- Cannot create wheel for version 0.14.2 on Windows HOT 1
- [BUG] Unable to load CLIPVisionModel parameters properly in Zero Stage 3 HOT 2
- [BUG] Can't pickle local object 'instrument_w_nvtx.<locals>.wrapped_fn' HOT 4
- [BUG] Multi-gpu training is much lower than single gpu (due to additional processes?)
- [REQUEST] Remove scary warnings from deepspeed import
- [BUG]output tensor must have the same type as input tensor in PPO training script of TRL HOT 4
- [REQUEST] How can one specify the CPU architecture to target. HOT 2
- how to gather checkpoints to master node during multi-nodes training HOT 6
- [REQUEST] i want to know how to merge deepspeed multi gpu optim file into one pytorch optim.pt file ? HOT 8
- [BUG] [Regression] Adam Offload Runtime Error with DeepSpeed v0.14.2 HOT 3
- [REQUEST] too many unrelated warning HOT 1
- [REQUEST] Use python sysconfig to generate CFLAGs HOT 1
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.