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-2_auto
results: []
Gemma-2-2B_task-2_60-samples_config-2_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: 0.6938
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: 16
- total_train_batch_size: 16
- 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 |
|---|---|---|---|
| 0.9552 | 0.6957 | 2 | 1.0746 |
| 0.9555 | 1.7391 | 5 | 1.0326 |
| 0.8712 | 2.7826 | 8 | 0.9149 |
| 0.7712 | 3.8261 | 11 | 0.8319 |
| 0.6958 | 4.8696 | 14 | 0.7876 |
| 0.6428 | 5.9130 | 17 | 0.7529 |
| 0.5843 | 6.9565 | 20 | 0.7262 |
| 0.561 | 8.0 | 23 | 0.7111 |
| 0.5211 | 8.6957 | 25 | 0.7022 |
| 0.456 | 9.7391 | 28 | 0.6938 |
| 0.4502 | 10.7826 | 31 | 0.6950 |
| 0.3993 | 11.8261 | 34 | 0.7011 |
| 0.3589 | 12.8696 | 37 | 0.7186 |
| 0.3157 | 13.9130 | 40 | 0.7445 |
| 0.2717 | 14.9565 | 43 | 0.7838 |
| 0.2344 | 16.0 | 46 | 0.8412 |
| 0.1863 | 16.6957 | 48 | 0.8866 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1