What dataset did you train this on? #3
Verified answer
by
Bhavesh716
nitnem4all
asked this question in
Q&A
-
|
Hey, just going through the repo — what kind of data did you use in data.txt to get decent results? Any specific recommendations for a beginner? |
Beta Was this translation helpful? Give feedback.
Admin verified this answer by
Bhavesh716
Apr 19, 2026
Replies: 1 comment
-
|
Hey! I trained it on a simple plain text file — anything works really, a book, some articles, or even Wikipedia dumps. For beginners I'd suggest starting with a single book or story (~1MB). Clean, consistent text gives the best results for a small model like this. Just drop it in data.txt and run train.py — that's it. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hey! I trained it on a simple plain text file — anything works really, a book, some articles, or even Wikipedia dumps.
For beginners I'd suggest starting with a single book or story (~1MB). Clean, consistent text gives the best results for a small model like this.
Just drop it in data.txt and run train.py — that's it.