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-128k
results: []
gemma2b-summarize-gemini1_5flash-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: 2.5119
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: 15
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.1289 | 1.0 | 208 | 2.5162 |
| 1.0298 | 2.0 | 416 | 2.4574 |
| 0.9905 | 3.0 | 624 | 2.4455 |
| 0.9668 | 4.0 | 832 | 2.4518 |
| 0.9507 | 5.0 | 1040 | 2.4578 |
| 0.9348 | 6.0 | 1248 | 2.4685 |
| 0.9236 | 7.0 | 1456 | 2.4789 |
| 0.9156 | 8.0 | 1664 | 2.4831 |
| 0.8987 | 9.0 | 1872 | 2.4963 |
| 0.9008 | 10.0 | 2080 | 2.5021 |
| 0.8976 | 11.0 | 2288 | 2.5050 |
| 0.8941 | 12.0 | 2496 | 2.5107 |
| 0.8878 | 13.0 | 2704 | 2.5123 |
| 0.8896 | 14.0 | 2912 | 2.5120 |
| 0.8797 | 15.0 | 3120 | 2.5119 |
Framework versions
- PEFT 0.11.1
- Transformers 4.40.1
- Pytorch 2.2.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1