flan-t5-base

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: 0.5164

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• 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
No log	1.0	7	0.5247
0.7817	2.0	14	0.5236
0.7751	3.0	21	0.5227
0.7751	4.0	28	0.5217
0.7577	5.0	35	0.5201
0.7611	6.0	42	0.5194
0.7611	7.0	49	0.5187
0.7587	8.0	56	0.5181
0.7491	9.0	63	0.5176
0.7497	10.0	70	0.5171
0.7497	11.0	77	0.5167
0.7639	12.0	84	0.5165
0.7687	13.0	91	0.5164
0.7687	14.0	98	0.5164
0.7907	15.0	105	0.5164

Framework versions

• PEFT 0.14.0
• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.0
• Tokenizers 0.21.0