Comments (4)
@carmocca we should be careful recommending such step methods. Currently, I would advise only going through forward() in Fabric. Due to the technical limitations of the ddp wrapper (which only wraps forward). The LM integration in Fabric has this issue Lightning-AI/pytorch-lightning#16856.
from lit-llama.
To be honest, I would just remove it if it's not used. There is no downside of removing it, and it would (slightly) aid the overall goal of having a minimal, readable implementation.
from lit-llama.
My bad, itβs a remnant from nanoGPT :-)
from lit-llama.
It could be used in the training scripts if we want to
from lit-llama.
Related Issues (20)
- Mistral Model HOT 1
- Adapter finetuning do not run on two cards (A100 40G)
- [question] error message while finetuning HOT 2
- Question about 'validating...' from lora.py
- [question] assert lora_path.is_file() error
- [question] nan loss value and run time error
- How to do conversation with fine tuned model?
- Running into StopIteration with single node multi GPU pretraining against the redpajama sample HOT 5
- Why is LLaMA response to queries in the conversation so wrong?
- RuntimeError: probability tensor contains either `inf`, `nan` or element < 0
- Full fine-tuning on Alpaca dataset with 4 L40s GPUs fails 8 hours into the training job with index_copy_ HOT 2
- How to convert hf weight of 70b to lit-lamma weights?
- How to quantize LLama in fine-tuning ?
- RuntimeError: cutlassF: no kernel found to launch! HOT 1
- RuntimeError: Expected x1.dtype() == cos.dtype() to be true, but got false. (Could this error message be improved? If so, please report an enhancement request to PyTorch.) HOT 1
- Can I use Lightning fabirc to pre train llama2 on v100?
- Ban some tokens
- Error: git submodule update --init --recursive -q did not run successfully HOT 1
- Beam search generation
- Issue with Rotary Embedding Initialization when the number of devices is > 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π 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. πππ
-
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 β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from lit-llama.