metadata
license: mit
library_name: peft
tags:
- generated_from_trainer
base_model: EleutherAI/gpt-neo-1.3B
model-index:
- name: peft-starcoder-lora-a100
results: []
peft-starcoder-lora-a100
This model is a fine-tuned version of EleutherAI/gpt-neo-1.3B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9114
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: 0.0005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 30
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.4496 | 0.05 | 100 | 0.9015 |
| 0.9852 | 0.1 | 200 | 0.8839 |
| 0.7642 | 0.15 | 300 | 0.8856 |
| 1.1261 | 0.2 | 400 | 0.8876 |
| 0.8212 | 0.25 | 500 | 0.8850 |
| 0.6402 | 0.3 | 600 | 0.8740 |
| 0.6944 | 0.35 | 700 | 0.8867 |
| 0.7385 | 0.4 | 800 | 0.8854 |
| 0.821 | 0.45 | 900 | 0.8824 |
| 0.5807 | 0.5 | 1000 | 0.8902 |
| 0.8403 | 0.55 | 1100 | 0.8986 |
| 0.7615 | 0.6 | 1200 | 0.8962 |
| 0.5245 | 0.65 | 1300 | 0.8974 |
| 0.7926 | 0.7 | 1400 | 0.9113 |
| 0.7051 | 0.75 | 1500 | 0.9064 |
| 0.6337 | 0.8 | 1600 | 0.9068 |
| 0.5369 | 0.85 | 1700 | 0.9126 |
| 0.7193 | 0.9 | 1800 | 0.9144 |
| 0.7216 | 0.95 | 1900 | 0.9111 |
| 0.4985 | 1.0 | 2000 | 0.9114 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1