Update README.md
Browse files
README.md
CHANGED
|
@@ -8,7 +8,7 @@ tags:
|
|
| 8 |
- absa
|
| 9 |
- qlora
|
| 10 |
---
|
| 11 |
-
#
|
| 12 |
|
| 13 |
## Model Details
|
| 14 |
|
|
@@ -52,7 +52,7 @@ This model is an aspect based sentiment analysis model fine-tuned from the Llama
|
|
| 52 |
- Ratio of steps for linear warmup: 0.03
|
| 53 |
- Group sequences into batches with the same length: True
|
| 54 |
- Save checkpoint every X update steps: 0 (disabled)
|
| 55 |
-
- Log every X update steps:
|
| 56 |
|
| 57 |
### SFT (Sequence-level Fine-Tuning)
|
| 58 |
- Maximum sequence length: Not specified
|
|
|
|
| 8 |
- absa
|
| 9 |
- qlora
|
| 10 |
---
|
| 11 |
+
# llama-2-7b-absa-semeval-2016
|
| 12 |
|
| 13 |
## Model Details
|
| 14 |
|
|
|
|
| 52 |
- Ratio of steps for linear warmup: 0.03
|
| 53 |
- Group sequences into batches with the same length: True
|
| 54 |
- Save checkpoint every X update steps: 0 (disabled)
|
| 55 |
+
- Log every X update steps: 100
|
| 56 |
|
| 57 |
### SFT (Sequence-level Fine-Tuning)
|
| 58 |
- Maximum sequence length: Not specified
|