metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: multi-emails-hq-pythia-410m-deduped-r1
results: []
multi-emails-hq-pythia-410m-deduped-r1
This model is a fine-tuned version of EleutherAI/pythia-410m-deduped on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.1018
- Accuracy: 0.6157
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: 8
- eval_batch_size: 2
- seed: 69
- distributed_type: multi-GPU
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 4.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.3067 | 1.0 | 38 | 2.3559 | 0.5594 |
| 1.9243 | 2.0 | 76 | 2.1283 | 0.5975 |
| 1.5971 | 3.0 | 114 | 2.0759 | 0.6140 |
| 1.4139 | 4.0 | 152 | 2.1018 | 0.6157 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 2.0.0.dev20230130+cu118
- Datasets 2.9.0
- Tokenizers 0.13.2