ll PRO
Etherll
AI & ML interests
None yet
Recent Activity
liked
a model
1 day ago
open-r1/OlympicCoder-7B
liked
a Space
22 days ago
Reality123b/XylariaDeepReason
reacted
to
s-emanuilov's
post
with 🔥
about 1 month ago
Tutorial 💥 Training a non-English reasoning model with GRPO and Unsloth
I wanted to share my experiment with training reasoning models in languages other than English/Chinese.
Using Llama 3.1 8B as base, GRPO trainer from trl, and Unsloth optimizations, I got a working prototype in Bulgarian after ~5 hours on an L40S GPU. The approach should work for any language where the base model has some pre-training coverage.
Full code and tutorial here: https://unfoldai.com/reasoning-in-a-non-english-language/
The model itself: https://huggingface.co/s-emanuilov/LLMBG-Llama-3.1-8B-BG-Reasoning-v0.1
I hope this helps anyone looking to build reasoning models in their language.
Organizations
Etherll's activity
Model benchmarks degraded after re-evaluation
1
#1018 opened 4 months ago
by
Etherll
Adding Evaluation Results
#1 opened 4 months ago
by
leaderboard-pr-bot
Upload 5 files
#1 opened 5 months ago
by
rombodawg
https://huggingface.co/Etherll/Herplete-LLM-Llama-3.1-8b
1
#313 opened 6 months ago
by
Etherll
Adding Evaluation Results
#2 opened 6 months ago
by
leaderboard-pr-bot
Adding Evaluation Results
#1 opened 6 months ago
by
leaderboard-pr-bot
[bot] Conversion to Parquet
#1 opened 7 months ago
by
parquet-converter
Create generation_config.json
#2 opened over 1 year ago
by
Etherll
Create generation_config.json
#1 opened over 1 year ago
by
Etherll
Create generation_config.json
#1 opened over 1 year ago
by
Etherll