Downloads · 30 days
67
6% of all-time downloads
ritwikm/gandhi-gpt
gandhi-gpt is a text generation model from ritwikm. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as gpl.
We have scrapped all the collected works of Mohandas Karamchand Gandhi (aka Mahatma Gandhi) from here (now archived here).
Downloads · 30 days
67
6% of all-time downloads
All-time downloads
1.1K
Public
Parameters
137M
1.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.bin510 MB · 44%
How the weights are stored.
F32124M · 91%
From the Hugging Face model README
We have scrapped all the collected works of Mohandas Karamchand Gandhi (aka Mahatma Gandhi) from here (now archived here).
Volume annotated all writings of Gandhi are available in code/gandhi-all.txt in this repository.
Then we cleaned the text so that it contains only the writings of Gandhi without footnotes, titles, and other texts.
We observed that (after cleaning), Gandhi wrote 755468 sentences.
We first fine-tuned gpt-2 for 1 epoch on the English corpus (after cleaning*) of Ai4Bharat.
Since the above dataset contains news regarding Indian subcontinent. We thought that with this fine-tuning, the model will get familiary with India specific terms.
Then we further fine-tuned this model on sentences written by Gandhi (for 3 epochs).
Here is the colab link with a working example.
*Before cleaning #sents = 54M, after cleaning 42M. We simply took those English sentences which ends with a full-stop.