---
base_model: ai-forever/rugpt3small_based_on_gpt2
tags:
- generated_from_trainer
model-index:
- name: rugpt3small
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# rugpt3small

This model is a fine-tuned version of [ai-forever/rugpt3small_based_on_gpt2](https://huggingface.co/ai-forever/rugpt3small_based_on_gpt2) on an unknown dataset.

## 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: 8
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 24
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3

### Training results



### Framework versions

- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Tokenizers 0.19.1