Edit model card

gpt2-XYZCompany-500-steps

This model is a question-answer chatbot for XYZCompany. It can answer questions related to the company. It is a fine-tuned version of gpt2 on XYZCompany's dataset containing question-answer pairs. It achieves the following results on the evaluation set:

  • Loss: 0.3300

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: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1
  • training_steps: 500

Training results

Training Loss Epoch Step Validation Loss
0.471 0.32 50 0.4135
0.4572 0.63 100 0.3736
0.3903 0.95 150 0.3574
0.3748 1.27 200 0.3474
0.3639 1.58 250 0.3413
0.3515 1.9 300 0.3366
0.3539 2.22 350 0.3337
0.3604 2.53 400 0.3319
0.3579 2.85 450 0.3305
0.3176 3.16 500 0.3300

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.1.2
  • Datasets 2.17.1
  • Tokenizers 0.13.2
Downloads last month
0
Inference API
This model can be loaded on Inference API (serverless).

Finetuned from