Update README.md
Browse files
README.md
CHANGED
|
@@ -29,6 +29,8 @@ I modified the dataset for uncensoring, deduping, cleaning, and quality.
|
|
| 29 |
|
| 30 |
I added Jon Durbin's excellent Airoboros dataset to increase creativity.
|
| 31 |
|
|
|
|
|
|
|
| 32 |
## Training
|
| 33 |
It took 5 days to train 3 epochs on 4x A100s using qLoRA and Axolotl
|
| 34 |
|
|
|
|
| 29 |
|
| 30 |
I added Jon Durbin's excellent Airoboros dataset to increase creativity.
|
| 31 |
|
| 32 |
+
I added a curated subset of Samantha and WizardLM data to train it for multi-turn conversation.
|
| 33 |
+
|
| 34 |
## Training
|
| 35 |
It took 5 days to train 3 epochs on 4x A100s using qLoRA and Axolotl
|
| 36 |
|