|
---
|
|
license: other
|
|
base_model: NowaBwagel0/llama-68m-oasst
|
|
tags:
|
|
- generated_from_trainer
|
|
model-index:
|
|
- name: llama-68m-oasst
|
|
results: []
|
|
---
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
should probably proofread and complete it, then remove this comment. -->
|
|
|
|
# llama-68m-oasst
|
|
|
|
This model is a fine-tuned version of [NowaBwagel0/llama-68m-oasst](https://huggingface.co/NowaBwagel0/llama-68m-oasst) on an unknown dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 4.6059
|
|
|
|
## Model description
|
|
|
|
More information needed
|
|
|
|
## Intended uses & limitations
|
|
|
|
More information needed
|
|
|
|
## Training and evaluation data
|
|
|
|
More information needed
|
|
|
|
## Training procedure
|
|
|
|
### Training hyperparameters
|
|
|
|
The following hyperparameters were used during training:
|
|
- learning_rate: 2e-05
|
|
- train_batch_size: 4
|
|
- eval_batch_size: 4
|
|
- seed: 42
|
|
- gradient_accumulation_steps: 2
|
|
- total_train_batch_size: 8
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
- lr_scheduler_type: linear
|
|
- num_epochs: 9
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss |
|
|
|:-------------:|:------:|:----:|:---------------:|
|
|
| 0.4908 | 0.9987 | 382 | 4.3758 |
|
|
| 0.4608 | 2.0 | 765 | 4.4247 |
|
|
| 0.4452 | 2.9987 | 1147 | 4.4661 |
|
|
| 0.4285 | 4.0 | 1530 | 4.5001 |
|
|
| 0.4172 | 4.9987 | 1912 | 4.5308 |
|
|
| 0.4093 | 6.0 | 2295 | 4.5598 |
|
|
| 0.3643 | 6.9987 | 2677 | 4.5825 |
|
|
| 0.3423 | 8.0 | 3060 | 4.6003 |
|
|
| 0.3522 | 8.9882 | 3438 | 4.6059 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.41.2
|
|
- Pytorch 2.2.2+cu121
|
|
- Datasets 2.20.0
|
|
- Tokenizers 0.19.1
|
|
|
