Comments (4)
I'm also hitting some CUDA out of memory errors on models + data that I expect to more easily fit on a 40GB A100 MiG.
I'm not familiar with the lit-llama codebase, so I'm not sure what's potentially different in lit-parrot but wanted to note that I'm seeing something similar.
from litgpt.
Do you still see this behaviour, and if so, can you share exactly the code you ran and the arguments passed?
from litgpt.
This is because LLaMA fine-tuning is hardcoded to use 256
max_seq_length:
https://github.com/Lightning-AI/lit-llama/blob/main/scripts/prepare_alpaca.py#L26
https://github.com/Lightning-AI/lit-llama/blob/main/finetune/adapter.py#L52
Whereas this repository is configured to use the longest sequence length in alpaca: 1037
. If you override it to 256
in https://github.com/Lightning-AI/lit-gpt/blob/main/finetune/adapter.py#L30, you should see the times match.
from litgpt.
Actually I was using the pretrain script, and I think the max token length is fixed in both lit-llama and lit-gpt?
from litgpt.
Related Issues (20)
- Killed when saving LoRA weights HOT 9
- Mistral 7B v0.2 HOT 6
- Compatible with local 8xH100 instead of cloud? HOT 4
- Add back support for longest sequence first
- Data prep in pretraining tutorial does not work HOT 3
- Rerun Gemma configs HOT 1
- Error when passing the config file "config_hub/finetune/llama-2-7b/lora.yaml" to the finetune command HOT 1
- Finetuning run times out at evaluation step on multiple devices HOT 4
- Python 3.12 HOT 2
- Implementation of RoPE HOT 3
- Is there any plan to add Megatron-LM support to enable lit-gpt to train extreme size model? HOT 2
- Can I train a model on 7900XT 4 cards? HOT 1
- Calculate loss at beginning and end of training
- Gradient clipping
- Something introduced a LoRA merging bug HOT 7
- Defaults fail on small block size for some models HOT 1
- LoRA model tokenizer configuration fails to load HOT 8
- TypeError: unsupported operand type(s) for -: 'float' and 'NoneType' HOT 7
- Gradients in GPT module of the finetuning/lora.py script are always zero HOT 6
- Explain how to pretrain on a custom dataset
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 litgpt.