Model save
Browse files
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# SolaceAI
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the None dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -42,9 +44,18 @@ The following hyperparameters were used during training:
|
|
| 42 |
- optimizer: Use adamw_bnb_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
- lr_scheduler_warmup_steps: 2
|
| 45 |
-
- training_steps:
|
| 46 |
- mixed_precision_training: Native AMP
|
| 47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
### Framework versions
|
| 49 |
|
| 50 |
- PEFT 0.14.0
|
|
|
|
| 15 |
# SolaceAI
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 3.0826
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 44 |
- optimizer: Use adamw_bnb_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
- lr_scheduler_warmup_steps: 2
|
| 47 |
+
- training_steps: 3000
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 54 |
+
| 12.5314 | 0.1692 | 1000 | 3.1412 |
|
| 55 |
+
| 12.3003 | 0.3385 | 2000 | 3.0979 |
|
| 56 |
+
| 12.4202 | 0.5077 | 3000 | 3.0826 |
|
| 57 |
+
|
| 58 |
+
|
| 59 |
### Framework versions
|
| 60 |
|
| 61 |
- PEFT 0.14.0
|