Update README.md
Browse files
README.md
CHANGED
|
@@ -14,7 +14,9 @@ Cost of this fine-tune is about $10 in electricity. It took me 3 tries to get it
|
|
| 14 |
Base model used for fine-tuning was 200k context Yi-34B-Llama model shared by larryvrh.
|
| 15 |
|
| 16 |
I had to change max_positional_embeddings in config.json and model_max_length to 4096 for training to start, otherwise I was OOMing straight away.
|
| 17 |
-
My first attempt had max_positional_embeddings set to 16384 and model_max_length set to 200000. This allowed fine-tuning to finish, but model was broken after applying LoRA and merging it.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Prompt Format
|
| 20 |
|
|
|
|
| 14 |
Base model used for fine-tuning was 200k context Yi-34B-Llama model shared by larryvrh.
|
| 15 |
|
| 16 |
I had to change max_positional_embeddings in config.json and model_max_length to 4096 for training to start, otherwise I was OOMing straight away.
|
| 17 |
+
My first attempt had max_positional_embeddings set to 16384 and model_max_length set to 200000. This allowed fine-tuning to finish, but that model was broken after applying LoRA and merging it. \
|
| 18 |
+
Please use my lesson and be careful when setting positional embeddings for training \
|
| 19 |
+
<b>This model is a third attempt with AEZAKMI v2 dataset and it works perfectly fine.</b>
|
| 20 |
|
| 21 |
## Prompt Format
|
| 22 |
|