sft_cml4

This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.6611

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: 0.0005
• train_batch_size: 3
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.1822	0.1200	200	3.9880
3.9009	0.2400	400	3.9248
3.7846	0.3599	600	3.9328
3.7095	0.4799	800	3.8393
3.5043	0.5999	1000	3.8130
3.4826	0.7199	1200	3.7608
3.3511	0.8398	1400	3.6997
3.3243	0.9598	1600	3.6349
2.5235	1.0798	1800	3.7826
2.0758	1.1998	2000	3.7627
2.0932	1.3197	2200	3.7233
1.9962	1.4397	2400	3.7059
2.0072	1.5597	2600	3.6756
1.9642	1.6797	2800	3.6801
1.9062	1.7996	3000	3.6658
1.9279	1.9196	3200	3.6611

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.2+cu118
• Datasets 2.19.0
• Tokenizers 0.19.1