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-gemini1_5flash-2k
results: []
gemma2b-summarize-gemini1_5flash-2k
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: 2.6449
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: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- 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 |
|---|---|---|---|
| 2.9846 | 0.8571 | 3 | 3.1614 |
| 2.8305 | 2.0 | 7 | 2.8875 |
| 2.3403 | 2.8571 | 10 | 2.8343 |
| 2.3403 | 4.0 | 14 | 2.7485 |
| 2.0675 | 4.8571 | 17 | 2.6913 |
| 1.8758 | 6.0 | 21 | 2.6605 |
| 1.8758 | 6.8571 | 24 | 2.6488 |
| 1.7659 | 8.0 | 28 | 2.6451 |
| 1.7298 | 8.5714 | 30 | 2.6449 |
Framework versions
- PEFT 0.11.1
- Transformers 4.40.1
- Pytorch 2.2.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1