Update README.md
Browse files
README.md
CHANGED
|
@@ -4,4 +4,17 @@ datasets:
|
|
| 4 |
- HuggingFaceH4/ultrachat_200k
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
- HuggingFaceH4/ultrachat_200k
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# Introduction
|
| 10 |
+
This model vistagi/Mixtral-8x7b-v0.1-sft is trained with Ultrachat-200K dataset through supervised finetuning using Mixtral-8x7b-v0.1 as the baseline model.
|
| 11 |
+
The training is done with bfloat16 precision using LoRA.
|
| 12 |
+
|
| 13 |
+
## Details
|
| 14 |
+
Used Librarys
|
| 15 |
+
- torch
|
| 16 |
+
- deepspeed
|
| 17 |
+
- pytorch lightning
|
| 18 |
+
- transformers
|
| 19 |
+
- peft
|
| 20 |
+
|