metadata
license: other
tags:
- generated_from_trainer
base_model: facebook/opt-125m
model-index:
- name: opt125_wiki_rlo_k50
results: []
opt125_wiki_rlo_k50
This model is a fine-tuned version of facebook/opt-125m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.8235
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.1416 | 0.8340 | 500 | 2.8959 |
| 2.9988 | 1.6681 | 1000 | 2.8387 |
| 2.8886 | 2.5021 | 1500 | 2.8256 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.2
- Datasets 2.19.1
- Tokenizers 0.19.1