|
--- |
|
license: gemma |
|
library_name: peft |
|
tags: |
|
- alignment-handbook |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
base_model: google/gemma-2b |
|
datasets: |
|
- llama-duo/synth_summarize_dataset_dedup |
|
model-index: |
|
- name: gemma2b-summarize-gpt4o-32k |
|
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. --> |
|
|
|
# gemma2b-summarize-gpt4o-32k |
|
|
|
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the llama-duo/synth_summarize_dataset_dedup dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.6191 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 3 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 48 |
|
- total_eval_batch_size: 24 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.3438 | 1.0 | 73 | 2.5419 | |
|
| 1.1767 | 2.0 | 146 | 2.4979 | |
|
| 1.1163 | 3.0 | 219 | 2.4967 | |
|
| 1.0605 | 4.0 | 292 | 2.5121 | |
|
| 1.0362 | 5.0 | 365 | 2.5401 | |
|
| 1.0052 | 6.0 | 438 | 2.5711 | |
|
| 0.9813 | 7.0 | 511 | 2.5912 | |
|
| 0.9593 | 8.0 | 584 | 2.6132 | |
|
| 0.953 | 9.0 | 657 | 2.6180 | |
|
| 0.9482 | 10.0 | 730 | 2.6191 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.11.1 |
|
- Transformers 4.41.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.2 |
|
- Tokenizers 0.19.1 |
