en_ru_translate_red

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

• Loss: 2.0132
• Bleu: 0.2679
• Gen Len: 19.3457

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: 2e-05
• train_batch_size: 100
• eval_batch_size: 100
• seed: 42
• optimizer: Use 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	140	2.2124	0.1262	19.4434
No log	2.0	280	2.1754	0.1343	19.3963
No log	3.0	420	2.1442	0.1992	19.3531
2.3611	4.0	560	2.1182	0.2263	19.3411
2.3611	5.0	700	2.0972	0.243	19.3446
2.3611	6.0	840	2.0786	0.2477	19.35
2.3611	7.0	980	2.0637	0.2395	19.3517
2.2675	8.0	1120	2.0509	0.2551	19.3443
2.2675	9.0	1260	2.0406	0.2641	19.3374
2.2675	10.0	1400	2.0321	0.2701	19.3409
2.2167	11.0	1540	2.0252	0.2673	19.3326
2.2167	12.0	1680	2.0198	0.2676	19.3409
2.2167	13.0	1820	2.0161	0.2671	19.3411
2.2167	14.0	1960	2.0140	0.2673	19.3429
2.1924	15.0	2100	2.0132	0.2679	19.3457

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0