arduino-fine_tuned

This model is a fine-tuned version of Salesforce/codegen-350M-multi on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2232

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5517	0.9964	140	0.3323
0.3111	2.0	281	0.2730
0.2641	2.9964	421	0.2525
0.2342	4.0	562	0.2399
0.2181	4.9964	702	0.2317
0.1998	6.0	843	0.2298
0.1897	6.9964	983	0.2242
0.1766	8.0	1124	0.2231
0.1694	8.9964	1264	0.2222
0.1606	9.9644	1400	0.2232

Framework versions

• PEFT 0.12.0
• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1