Comments (1)
Thanks for your interest in LMFlow! We have fixed several bugs of the LISA implementation in LMFlow, it would be nice if you could check whether the implementation matches our latest version.
If the implementation is correct, it is worth trying:
- A smaller
lisa_interval_steps
, like 5 instead of 50, since more frequent sampling allows more layers to be covered. - If that doesn't work, you may try larger
lisa_activated_layers
. We have observed that in some cases, such as llama-2-70b, deeper structure requires largerlisa_activated_layers
. This may be also the case when data distribution is not that easy to learn.
Hope this information can be helpful. Please feel free to let us know if further problems are encountered 😄
from lmflow.
Related Issues (20)
- Causal LM finetuning HOT 3
- ValueError: mutable default <class 'lmflow.utils.conversation_formatter.StringFormatter'> for field user_formatter is not allowed: use default_factory HOT 8
- Hello , Can LMFlow support Qwen1.5-1.8B model Fine-tuning? HOT 3
- LMFlow not support NVIDIA driver 11070? HOT 2
- Hello,Where is the script run_finetune_with_lora_save_aggregated_weights.sh?Why I can't find it in LMFlow/scripts ? HOT 2
- Out Of Memory Issue LISA HOT 4
- Hello,How to go on fine-tuning with checkpoint?
- Hello,How to go on fine-tuning with checkpoint?
- Hello,How to go on fine-tuning with checkpoint? HOT 2
- [BUG] did not output the eval results at all. HOT 3
- Fine-Tuning Crashes for no reason when Eight GPU cards are used. HOT 4
- Hello, why I fine-tuning Qwen1.5-1.8B-Base and test with CMMLU, the model answer repetition HOT 6
- Add 'validation_split_percentage' and 'evaluation_strategy' parameters for Trainers HOT 1
- [BUG] Can Lisa be used for chatglm3 with lmflow? HOT 1
- Weird Loss Curve HOT 1
- Discussion about LISA HOT 1
- [BUG]when map the dataset, i set the num_proc = 2 or 4, it will make mistakes. HOT 8
- Training was successful on a single card 4090GPU, but an error was reported on a 3*4090GPU. why HOT 1
- Full parameter fine-tuning cannot be trained 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 lmflow.