Update README.md
Browse files
README.md
CHANGED
|
@@ -22,6 +22,7 @@ pipeline_tag: text-generation
|
|
| 22 |
|
| 23 |
# Information
|
| 24 |
Advanced, high-quality and **lite** reasoning for a tiny size that you can run on your phone.
|
|
|
|
| 25 |
At original quality, it runs at ~400 tokens/second on a single H100 Nvidia GPU from Friendli.
|
| 26 |
|
| 27 |
Trained similarly to Deepseek R1, we used Smollm2 as a base model, then we've SFT fine tuned on reasoning using our own private superthoughts instruct dataset & modified the tokenizer slightly, after the SFT fine tuning we used GRPO to further amplify it's mathematics & problem solving abilities.
|
|
|
|
| 22 |
|
| 23 |
# Information
|
| 24 |
Advanced, high-quality and **lite** reasoning for a tiny size that you can run on your phone.
|
| 25 |
+
|
| 26 |
At original quality, it runs at ~400 tokens/second on a single H100 Nvidia GPU from Friendli.
|
| 27 |
|
| 28 |
Trained similarly to Deepseek R1, we used Smollm2 as a base model, then we've SFT fine tuned on reasoning using our own private superthoughts instruct dataset & modified the tokenizer slightly, after the SFT fine tuning we used GRPO to further amplify it's mathematics & problem solving abilities.
|