metadata
base_model: google/gemma-2-2b-it
datasets:
- GaetanMichelet/chat-60_ft_task-2_auto
library_name: peft
license: gemma
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: Gemma-2-2B_task-2_60-samples_config-1_full_auto
results: []
Gemma-2-2B_task-2_60-samples_config-1_full_auto
This model is a fine-tuned version of google/gemma-2-2b-it on the GaetanMichelet/chat-60_ft_task-2_auto dataset. It achieves the following results on the evaluation set:
- Loss: 0.8903
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.2771 | 0.8696 | 5 | 1.3014 |
| 1.2877 | 1.9130 | 11 | 1.2201 |
| 1.1635 | 2.9565 | 17 | 1.1144 |
| 1.0236 | 4.0 | 23 | 1.0385 |
| 0.9255 | 4.8696 | 28 | 0.9711 |
| 0.8522 | 5.9130 | 34 | 0.9287 |
| 0.7873 | 6.9565 | 40 | 0.9061 |
| 0.7746 | 8.0 | 46 | 0.8943 |
| 0.7645 | 8.8696 | 51 | 0.8906 |
| 0.7497 | 9.9130 | 57 | 0.8903 |
| 0.677 | 10.9565 | 63 | 0.8981 |
| 0.6337 | 12.0 | 69 | 0.9132 |
| 0.637 | 12.8696 | 74 | 0.9355 |
| 0.5524 | 13.9130 | 80 | 0.9835 |
| 0.5137 | 14.9565 | 86 | 1.0106 |
| 0.5166 | 16.0 | 92 | 1.0932 |
| 0.3388 | 16.8696 | 97 | 1.1325 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1