metadata
license: apache-2.0
base_model: EleutherAI/pythia-2.8b
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: jeffrey-walker-teal
results: []
jeffrey-walker-teal
This model is a fine-tuned version of EleutherAI/pythia-2.8b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 7.8984
- Accuracy: 0.4
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: 1e-05
- train_batch_size: 4
- eval_batch_size: 16
- seed: 24
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 2500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.1422 | 8.5561 | 400 | 4.6875 | 0.4 |
| 0.0121 | 17.1123 | 800 | 6.6641 | 0.5 |
| 0.0 | 25.6684 | 1200 | 4.7148 | 0.8 |
| 0.028 | 34.2246 | 1600 | 6.6992 | 0.7 |
| 0.0041 | 42.7807 | 2000 | 7.4570 | 0.4 |
| 0.0021 | 51.3369 | 2400 | 7.8984 | 0.4 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.1.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1