|
--- |
|
license: apache-2.0 |
|
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0 |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: Tukan-1.1B-Chat-reasoning-sft-COLA-2.5epr-R8-A16 |
|
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. --> |
|
|
|
# Tukan-1.1B-Chat-reasoning-sft-COLA-2.5epr-R8-A16 |
|
|
|
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) on the generator dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0216 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 6 |
|
- eval_batch_size: 3 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- gradient_accumulation_steps: 20 |
|
- total_train_batch_size: 120 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- num_epochs: 2.5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.1073 | 0.24 | 10 | 1.1011 | |
|
| 1.1024 | 0.47 | 20 | 1.0842 | |
|
| 1.0961 | 0.71 | 30 | 1.0675 | |
|
| 1.066 | 0.94 | 40 | 1.0529 | |
|
| 1.0598 | 1.18 | 50 | 1.0413 | |
|
| 1.0384 | 1.42 | 60 | 1.0326 | |
|
| 1.0356 | 1.65 | 70 | 1.0268 | |
|
| 1.0378 | 1.89 | 80 | 1.0235 | |
|
| 1.0376 | 2.12 | 90 | 1.0220 | |
|
| 1.0309 | 2.36 | 100 | 1.0215 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.2.0a0+gitd925d94 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.15.0 |
|
|
