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-128k
results: []
gemma2b-summarize-claude3sonnet-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: 2.5270
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 |
|---|---|---|---|
| 1.097 | 1.0 | 402 | 2.4839 |
| 1.0176 | 2.0 | 804 | 2.4534 |
| 0.9746 | 3.0 | 1206 | 2.4625 |
| 0.9525 | 4.0 | 1608 | 2.4586 |
| 0.9361 | 5.0 | 2010 | 2.4669 |
| 0.9077 | 6.0 | 2412 | 2.4844 |
| 0.896 | 7.0 | 2814 | 2.4947 |
| 0.8858 | 8.0 | 3216 | 2.5056 |
| 0.8811 | 9.0 | 3618 | 2.5128 |
| 0.8634 | 10.0 | 4020 | 2.5166 |
| 0.8758 | 11.0 | 4422 | 2.5237 |
| 0.8644 | 12.0 | 4824 | 2.5264 |
| 0.8641 | 13.0 | 5226 | 2.5268 |
| 0.8607 | 14.0 | 5628 | 2.5271 |
| 0.8609 | 15.0 | 6030 | 2.5270 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1