microsoft_CodeGPT-small-java_1_ft_clm

This model is a fine-tuned version of microsoft/CodeGPT-small-java on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2841
• Accuracy: 0.7565

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: 8e-05
• train_batch_size: 8
• eval_batch_size: 24
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.02
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6967	0.5028	500	1.3162	0.7521
0.6481	1.0050	1000	1.3002	0.7551
0.6091	1.5078	1500	1.2965	0.7550
0.6091	2.0101	2000	1.2894	0.7556
0.5805	2.5128	2500	1.2918	0.7556
0.557	3.0151	3000	1.2862	0.7559
0.5513	3.5178	3500	1.2841	0.7565
0.5468	4.0201	4000	1.2849	0.7562
0.5513	4.5229	4500	1.2895	0.7559

Framework versions

• Transformers 4.53.0
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.21.2