README.md

metadata

tags:
  - generated_from_trainer
datasets:
  - roneneldan/TinyStories
metrics:
  - accuracy
model-index:
  - name: gpt2_u010_tiny-stories_1024_dpos
    results:
      - task:
          name: Causal Language Modeling
          type: text-generation
        dataset:
          name: roneneldan/TinyStories
          type: roneneldan/TinyStories
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6790614246318416

gpt2_u010_tiny-stories_1024_dpos

This model is a fine-tuned version of on the roneneldan/TinyStories dataset. It achieves the following results on the evaluation set:

• Loss: 1.2086
• Accuracy: 0.6791

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: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.9081	0.0523	1000	2.4416	0.4491
1.966	0.1046	2000	1.7895	0.5706
1.7221	0.1570	3000	1.6081	0.6024
1.6057	0.2093	4000	1.5104	0.6201
1.5313	0.2616	5000	1.4469	0.6323
1.4786	0.3139	6000	1.4022	0.6407
1.439	0.3662	7000	1.3702	0.6465
1.4105	0.4186	8000	1.3421	0.6523
1.3836	0.4709	9000	1.3161	0.6571
1.3647	0.5232	10000	1.2970	0.6609
1.3473	0.5755	11000	1.2799	0.6645
1.3306	0.6278	12000	1.2652	0.6672
1.3141	0.6802	13000	1.2522	0.6699
1.3002	0.7325	14000	1.2418	0.6720
1.2918	0.7848	15000	1.2328	0.6740
1.2841	0.8371	16000	1.2239	0.6757
1.2781	0.8894	17000	1.2175	0.6771
1.271	0.9418	18000	1.2123	0.6782
1.2651	0.9941	19000	1.2088	0.6790

Framework versions

• Transformers 4.42.3
• Pytorch 2.2.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1