metadata
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-128k
results: []
gemma2b-summarize-gpt4o-128k
This model is a fine-tuned version of google/gemma-2b on the llama-duo/synth_summarize_dataset_dedup dataset. It achieves the following results on the evaluation set:
- Loss: 2.5233
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: 15
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.2085 | 1.0 | 293 | 2.4863 |
| 1.1135 | 2.0 | 586 | 2.4516 |
| 1.0715 | 3.0 | 879 | 2.4473 |
| 1.0471 | 4.0 | 1172 | 2.4524 |
| 1.0357 | 5.0 | 1465 | 2.4685 |
| 0.993 | 6.0 | 1758 | 2.4703 |
| 0.9941 | 7.0 | 2051 | 2.4906 |
| 0.9844 | 8.0 | 2344 | 2.4896 |
| 0.9779 | 9.0 | 2637 | 2.5025 |
| 0.9639 | 10.0 | 2930 | 2.5126 |
| 0.952 | 11.0 | 3223 | 2.5192 |
| 0.9505 | 12.0 | 3516 | 2.5205 |
| 0.9442 | 13.0 | 3809 | 2.5223 |
| 0.9469 | 14.0 | 4102 | 2.5227 |
| 0.9444 | 15.0 | 4395 | 2.5233 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1