metadata
license: mit
library_name: peft
tags:
- trl
- sft
- conversation
base_model: openai-community/gpt2-medium
model-index:
- name: openai_gpt
results: []
datasets:
- daily_dialog
pipeline_tag: text-generation
openai_gpt
This model is a fine-tuned version of openai-community/gpt2-medium on the daily_dialog dataset.
Model description
Intended uses & limitations
The model can be used for friendly conversations
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0025
- train_batch_size: 3
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 6
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 200
Training results
- Training Loss : 2.113900
Framework versions
- PEFT 0.9.1.dev0
- Transformers 4.38.2
- Pytorch 2.0.0
- Datasets 2.18.0
- Tokenizers 0.15.2