03a7666349e52bd84dd031bf5597eb5a

This model is a fine-tuned version of google-t5/t5-large on the Helsinki-NLP/opus_books [es-ru] dataset. It achieves the following results on the evaluation set:

  • Loss: 1.2026
  • Data Size: 1.0
  • Epoch Runtime: 185.3567
  • Bleu: 6.0015

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: 8
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • total_train_batch_size: 32
  • total_eval_batch_size: 32
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: constant
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss Data Size Epoch Runtime Bleu
No log 0 0 3.6385 0 13.0732 0.3179
No log 1 419 2.4707 0.0078 14.7994 0.4113
No log 2 838 2.0728 0.0156 16.9704 0.7224
0.066 3 1257 1.9499 0.0312 20.6334 0.7431
0.066 4 1676 1.8554 0.0625 26.4972 0.6377
0.1259 5 2095 1.7783 0.125 37.3913 1.2844
0.2527 6 2514 1.6944 0.25 56.9756 1.8325
1.7354 7 2933 1.5901 0.5 102.7781 2.3839
1.5782 8.0 3352 1.4781 1.0 184.7438 2.9741
1.4939 9.0 3771 1.4056 1.0 185.4884 3.5832
1.4426 10.0 4190 1.3526 1.0 179.4644 3.7820
1.3493 11.0 4609 1.3130 1.0 175.0325 4.1739
1.3145 12.0 5028 1.2849 1.0 176.5448 4.3942
1.2422 13.0 5447 1.2626 1.0 177.8109 4.6982
1.2292 14.0 5866 1.2432 1.0 175.2749 5.0353
1.159 15.0 6285 1.2246 1.0 177.3543 5.1940
1.1365 16.0 6704 1.2178 1.0 173.9226 5.0895
1.1087 17.0 7123 1.2090 1.0 184.5408 5.3373
1.0513 18.0 7542 1.2023 1.0 180.8638 5.4654
1.0269 19.0 7961 1.2005 1.0 191.6813 5.4241
0.9834 20.0 8380 1.1986 1.0 184.3451 5.6317
0.963 21.0 8799 1.1956 1.0 178.4502 5.7080
0.9317 22.0 9218 1.1870 1.0 183.7618 5.7567
0.9179 23.0 9637 1.1981 1.0 179.5555 5.9397
0.8723 24.0 10056 1.1981 1.0 176.1049 5.9674
0.8653 25.0 10475 1.2113 1.0 180.2787 5.9325
0.8227 26.0 10894 1.2026 1.0 185.3567 6.0015

Framework versions

  • Transformers 4.57.0
  • Pytorch 2.8.0+cu128
  • Datasets 4.2.0
  • Tokenizers 0.22.1
Downloads last month
-
Safetensors
Model size
0.8B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for contemmcm/03a7666349e52bd84dd031bf5597eb5a

Finetuned
(171)
this model