e296c58d-f55a-456d-846b-a20f5752ff65

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: 0.5308

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: 0.00021
• 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	0.0005	1	1.4149
0.6033	0.0237	50	0.7537
0.5289	0.0473	100	0.6691
0.5413	0.0710	150	0.6276
0.505	0.0946	200	0.6103
0.4802	0.1183	250	0.5883
0.4843	0.1419	300	0.5656
0.455	0.1656	350	0.5479
0.4646	0.1892	400	0.5372
0.4644	0.2129	450	0.5313
0.4401	0.2365	500	0.5308

Framework versions

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