Commit
•
c4856c5
1
Parent(s):
c47853c
will update checkpoints everyday.
Browse files
README.md
CHANGED
|
@@ -13,6 +13,7 @@ ATM I'm using 2023.05.04v0 of the dataset and training full context.
|
|
| 13 |
# Notes:
|
| 14 |
So im only training 1 epoch as full context 30b takes a long time to train.
|
| 15 |
My 1 epoch will take me 8 days lol but lucly the LoRA feels fully functinal at epoch 1 as shown on my 13b one.
|
|
|
|
| 16 |
|
| 17 |
# How to test?
|
| 18 |
1. Download LLaMA-30B-HF: https://huggingface.co/Neko-Institute-of-Science/LLaMA-30B-HF
|
|
|
|
| 13 |
# Notes:
|
| 14 |
So im only training 1 epoch as full context 30b takes a long time to train.
|
| 15 |
My 1 epoch will take me 8 days lol but lucly the LoRA feels fully functinal at epoch 1 as shown on my 13b one.
|
| 16 |
+
Also I will be uploading checkpoints almost everyday.
|
| 17 |
|
| 18 |
# How to test?
|
| 19 |
1. Download LLaMA-30B-HF: https://huggingface.co/Neko-Institute-of-Science/LLaMA-30B-HF
|