---
library_name: transformers
license: mit
base_model: gpt2-medium
tags:
- generated_from_trainer
model-index:
- name: chessgpt2-medium-l
  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. -->

# chessgpt2-medium-l

This model is a fine-tuned version of [gpt2-medium](https://huggingface.co/gpt2-medium) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8574

## 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: 5e-05
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.025
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step  | Validation Loss |
|:-------------:|:------:|:-----:|:---------------:|
| 2.1756        | 0.1280 | 2000  | 1.5178          |
| 1.4207        | 0.2560 | 4000  | 1.2728          |
| 1.265         | 0.3839 | 6000  | 1.1726          |
| 1.1852        | 0.5119 | 8000  | 1.1138          |
| 1.1321        | 0.6399 | 10000 | 1.0720          |
| 1.0931        | 0.7679 | 12000 | 1.0384          |
| 1.0621        | 0.8958 | 14000 | 1.0097          |
| 1.0319        | 1.0238 | 16000 | 0.9891          |
| 0.9981        | 1.1518 | 18000 | 0.9680          |
| 0.9816        | 1.2798 | 20000 | 0.9498          |
| 0.9655        | 1.4077 | 22000 | 0.9360          |
| 0.951         | 1.5357 | 24000 | 0.9219          |
| 0.9386        | 1.6637 | 26000 | 0.9094          |
| 0.9264        | 1.7917 | 28000 | 0.8974          |
| 0.9154        | 1.9196 | 30000 | 0.8876          |
| 0.8973        | 2.0476 | 32000 | 0.8804          |
| 0.8734        | 2.1756 | 34000 | 0.8743          |
| 0.8684        | 2.3036 | 36000 | 0.8683          |
| 0.8638        | 2.4315 | 38000 | 0.8647          |
| 0.86          | 2.5595 | 40000 | 0.8611          |
| 0.8581        | 2.6875 | 42000 | 0.8589          |
| 0.8557        | 2.8155 | 44000 | 0.8579          |
| 0.8556        | 2.9434 | 46000 | 0.8574          |


### Framework versions

- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1