baseline_1-seed_123

This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.1634
• Precision: 0.6346
• Recall: 0.6920
• F1: 0.6620

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 123
• 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: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
0.5175	1.0	447	0.7010	0.6054	0.5725	0.5885
0.4308	2.0	894	1.0253	0.5945	0.7065	0.6457
0.417	3.0	1341	1.0984	0.5971	0.7464	0.6634
0.3336	4.0	1788	1.6736	0.5952	0.7138	0.6491
0.2283	5.0	2235	1.6788	0.6133	0.6667	0.6389
0.1776	6.0	2682	1.5558	0.5964	0.7174	0.6513
0.2152	7.0	3129	2.2730	0.6228	0.6341	0.6284
0.1292	8.0	3576	2.4796	0.6179	0.6268	0.6223
0.1486	9.0	4023	2.3853	0.6131	0.6775	0.6437
0.1032	10.0	4470	2.4573	0.6098	0.7246	0.6623
0.0678	11.0	4917	3.0114	0.6319	0.6594	0.6454
0.0335	12.0	5364	3.2091	0.6287	0.6993	0.6621
0.0627	13.0	5811	3.1270	0.6186	0.6993	0.6565
0.0814	14.0	6258	3.1125	0.6242	0.7101	0.6644
0.003	15.0	6705	3.1634	0.6346	0.6920	0.6620

Framework versions

• Transformers 4.52.0.dev0
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1