ielts_qg_flan_t5

This model is a fine-tuned version of t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3128
• Pred Len: 48.5714

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Pred Len
6.0427	1.0	195	4.1822	175.3802
3.387	2.0	390	2.7273	55.5741
2.8707	3.0	585	2.5263	42.6597
2.6846	4.0	780	2.4502	46.0989
2.556	5.0	975	2.4119	43.2357
2.5356	6.0	1170	2.3773	42.7395
2.5352	7.0	1365	2.3476	42.8023
2.5161	8.0	1560	2.3270	44.8441
2.424	9.0	1755	2.3197	40.2909
2.398	10.0	1950	2.3068	45.6863
2.4254	11.0	2145	2.2972	43.7034
2.36	12.0	2340	2.2903	45.0627
2.3012	13.0	2535	2.2876	42.5399
2.311	14.0	2730	2.2814	44.3479
2.2949	15.0	2925	2.2791	45.0760
2.2686	16.0	3120	2.2766	47.6160
2.2667	17.0	3315	2.2724	49.6996
2.2149	18.0	3510	2.2720	49.5817
2.2721	19.0	3705	2.2700	49.0989
2.2717	20.0	3900	2.2708	50.1901

Framework versions

• Transformers 4.53.3
• Pytorch 2.6.0+cu124
• Datasets 4.1.1
• Tokenizers 0.21.2