---
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-gemini1_5flash-64k
  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-gemini1_5flash-64k

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: 2.5156

## 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: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- total_eval_batch_size: 64
- 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 |
|:-------------:|:------:|:----:|:---------------:|
| 1.3387        | 0.9905 | 52   | 2.5711          |
| 1.0869        | 2.0    | 105  | 2.4883          |
| 1.0363        | 2.9905 | 157  | 2.4850          |
| 0.9967        | 4.0    | 210  | 2.4809          |
| 0.98          | 4.9905 | 262  | 2.4904          |
| 0.9574        | 6.0    | 315  | 2.4995          |
| 0.9392        | 6.9905 | 367  | 2.5062          |
| 0.9415        | 8.0    | 420  | 2.5143          |
| 0.9367        | 8.9905 | 472  | 2.5155          |
| 0.9348        | 9.9048 | 520  | 2.5156          |


### Framework versions

- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1