324b02c9-8e6a-4627-969b-a20d6fdafe80

This model is a fine-tuned version of jingyeom/seal3.1.6n_7b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0550

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: 0.000212
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0000	1	2.0472
0.1009	0.0016	50	0.1092
0.0839	0.0031	100	0.1910
0.1454	0.0047	150	0.0942
0.0771	0.0063	200	0.1272
0.0834	0.0078	250	0.0811
0.0811	0.0094	300	0.0691
0.0671	0.0110	350	0.0615
0.0671	0.0125	400	0.0587
0.0633	0.0141	450	0.0553
0.0673	0.0157	500	0.0550

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1