Merge branch 'main' of hf.co:evilfreelancer/ruGPT-3.5-13B-lora
Browse files
README.md
CHANGED
|
@@ -27,7 +27,7 @@ tags:
|
|
| 27 |
|
| 28 |
Welcome to the adapter-only version of ruGPT-3.5 13B LoRA. This model is built upon the foundation of [ruGPT-3.5-13B](https://huggingface.co/ai-forever/ruGPT-3.5-13B).
|
| 29 |
|
| 30 |
-
π Important: This model was trained using settings identical to [GigaSaiga](https://huggingface.co/IlyaGusev/gigasaiga_lora), but incorporates
|
| 31 |
|
| 32 |
π Training code is [here](https://github.com/EvilFreelancer/ruGPT-3.5-13B-lora).
|
| 33 |
|
|
|
|
| 27 |
|
| 28 |
Welcome to the adapter-only version of ruGPT-3.5 13B LoRA. This model is built upon the foundation of [ruGPT-3.5-13B](https://huggingface.co/ai-forever/ruGPT-3.5-13B).
|
| 29 |
|
| 30 |
+
π Important: This model was trained using settings identical to [GigaSaiga](https://huggingface.co/IlyaGusev/gigasaiga_lora), but incorporates additional dataset.
|
| 31 |
|
| 32 |
π Training code is [here](https://github.com/EvilFreelancer/ruGPT-3.5-13B-lora).
|
| 33 |
|