chrisociepa / allamo Goto Github PK
View Code? Open in Web Editor NEWSimple, hackable and fast implementation for training/finetuning medium-sized LLaMA-based models
License: MIT License
Simple, hackable and fast implementation for training/finetuning medium-sized LLaMA-based models
License: MIT License
I've trained a small model for hungarian language for 5 days, text generation is working well. Is it possible to use this model for text classification too? (after finetraining on 100k text+class pairs dataset somehow)
Where/how can be the classification hidden layer added/connected to this model?
line 137:
import_tokenizer(args.input_tokenizer_path, args.output_data_dir)
should be:
import_tokenizer(args.input_tokenizer_path, args.output_data_dir, args.max_block_size)
hello, thanks for the great work
i know that sample.py
, sample_api.py
exist but i just want to use the model in a standalone python script
because of the way AllamoConfiguration.__post_init__()
is defined, i cannot create a new instance AllamoConfiguration
to use it
is there any way to do it properly without touching the source code?
many thanks
Like example datasets and configs.
Could you please give more details on how to finetune llama on my own QA data?
Hey @chrisociepa, awesome repo! Could you pls shed some light on how the train.txt
file should look like?
Thank you!
It would be valuable to provide at least one sample config file...
Much larger universe of people who can use truly open Llama variants such as the two models above. Limiting the code to Facebook-approved academics severely limits its value and reach.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.