metadata
base_model: TinyLlama/TinyLlama_v1.1
datasets:
- generator
library_name: peft
license: apache-2.0
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: train_output_dir
results: []
train_output_dir
This model is a fine-tuned version of TinyLlama/TinyLlama_v1.1 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 1.5212
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9653 | 0.1575 | 20 | 1.9348 |
| 1.853 | 0.3150 | 40 | 1.7662 |
| 1.6943 | 0.4724 | 60 | 1.6475 |
| 1.6361 | 0.6299 | 80 | 1.5510 |
| 1.6067 | 0.7874 | 100 | 1.5212 |
Framework versions
- PEFT 0.11.1
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1