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-64k
results: []
gemma2b-summarize-gpt4o-64k
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: 2.5990
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 |
|---|---|---|---|
| 1.2959 | 1.0 | 146 | 2.5295 |
| 1.1524 | 2.0 | 292 | 2.4913 |
| 1.1138 | 3.0 | 438 | 2.4847 |
| 1.0703 | 4.0 | 584 | 2.4927 |
| 1.0423 | 5.0 | 730 | 2.5080 |
| 1.0322 | 6.0 | 876 | 2.5202 |
| 1.0113 | 7.0 | 1022 | 2.5385 |
| 0.9857 | 8.0 | 1168 | 2.5522 |
| 0.9865 | 9.0 | 1314 | 2.5657 |
| 0.9691 | 10.0 | 1460 | 2.5774 |
| 0.952 | 11.0 | 1606 | 2.5889 |
| 0.97 | 12.0 | 1752 | 2.5957 |
| 0.9514 | 13.0 | 1898 | 2.5988 |
| 0.9469 | 14.0 | 2044 | 2.5997 |
| 0.9469 | 15.0 | 2190 | 2.5990 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1