formula-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: 0.7278
• Exact Match: 0.029
• Sketch Match: 0.729

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: 5e-05
• train_batch_size: 48
• eval_batch_size: 48
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 192
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 0.05
• num_epochs: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Exact Match	Validation Loss	Sketch Match
28.3162	1.0	491	0.0	2.0387	0.4857
12.1008	2.0	982	0.0127	1.1106	0.6797
9.0784	3.0	1473	0.014	0.9603	0.7123
8.0024	4.0	1964	0.0207	0.8636	0.724
7.2896	5.0	2455	0.0217	0.8055	0.7257
6.8783	6.0	2946	0.026	0.7614	0.7237
6.6384	7.0	3437	0.7369	0.0283	0.7337
6.5362	8.0	3928	0.7278	0.029	0.729

Framework versions

• Transformers 5.2.0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2