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-claude3sonnet-64k
results: []
gemma2b-summarize-claude3sonnet-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: 2.5427
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 |
|---|---|---|---|
| 1.163 | 0.9975 | 200 | 2.5127 |
| 1.0647 | 2.0 | 401 | 2.4643 |
| 1.0051 | 2.9975 | 601 | 2.4610 |
| 0.9807 | 4.0 | 802 | 2.4767 |
| 0.9508 | 4.9975 | 1002 | 2.4788 |
| 0.9256 | 6.0 | 1203 | 2.4912 |
| 0.9216 | 6.9975 | 1403 | 2.5038 |
| 0.9094 | 8.0 | 1604 | 2.5124 |
| 0.8961 | 8.9975 | 1804 | 2.5246 |
| 0.8816 | 10.0 | 2005 | 2.5342 |
| 0.8722 | 10.9975 | 2205 | 2.5346 |
| 0.8768 | 12.0 | 2406 | 2.5410 |
| 0.8694 | 12.9975 | 2606 | 2.5415 |
| 0.8709 | 14.0 | 2807 | 2.5418 |
| 0.8781 | 14.9626 | 3000 | 2.5427 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1