metadata

license: mit
base_model: gpt2
tags:
  - generated_from_trainer
model-index:
  - name: GPT2-124M-wikitext-v0.1
    results: []
datasets:
  - wikitext
pipeline_tag: text-generation
co2_eq_emissions:
  emissions: 500
  training_type: fine-tuning
  source: mlco2
  geographical_location: Bucharest, Romania
  hardware_used: 1 x RTX 4090 GPU

🧠 GPT2-124M-wikitext-v0.1

This model is a fine-tuned version of gpt2 on the wikitext. It achieves the following results on the evaluation set:

Loss: 2.9841

Model description

This is a practical hands-on experience for better understanding 🤗 Transformers and 🤗 Datasets. This model is GPT2(124M) fine-tuned on wikitext(103-raw-v1) on 1 x RTX 4090.

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: 2e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
3.1335	1.0	57467	3.0363
3.0643	2.0	114934	2.9968
3.0384	3.0	172401	2.9841

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0