m2m100_418M-lecturaFacil

This model is a fine-tuned version of facebook/m2m100_418M on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.0661
  • Rouge1: 43.385
  • Rouge2: 32.8564
  • Rougel: 33.7117
  • Rougelsum: 42.0926

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: 5.6e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum
No log 1.0 126 1.7497 36.685 26.6869 30.2571 35.0041
2.0989 2.0 252 1.3739 40.2068 30.7895 33.2676 38.7904
2.0989 3.0 378 1.2066 43.2475 34.7388 36.3563 41.9648
1.217 4.0 504 1.0995 41.3675 31.9862 34.2429 40.0616
1.217 5.0 630 1.0401 41.7939 31.5659 33.7242 40.4383
0.8874 6.0 756 0.9996 42.1276 33.2102 35.023 40.7052
0.8874 7.0 882 0.9638 42.0925 32.145 34.0498 40.6997
0.6815 8.0 1008 0.9428 43.2718 34.0356 35.7494 41.9982
0.6815 9.0 1134 0.9420 43.4868 34.301 35.5118 42.3057
0.5505 10.0 1260 0.9328 41.9763 32.0642 33.8328 40.6177
0.5505 11.0 1386 0.9290 43.7249 34.629 35.8741 42.4228
0.4493 12.0 1512 0.9381 43.2557 33.8024 35.1489 41.9791
0.4493 13.0 1638 0.9413 41.7681 30.7291 31.8306 40.3795
0.3818 14.0 1764 0.9471 42.3205 31.4222 32.7127 40.8971
0.3818 15.0 1890 0.9592 43.3617 33.3281 34.1906 42.0589
0.3123 16.0 2016 0.9687 42.9163 32.1448 33.1292 41.5772
0.3123 17.0 2142 0.9803 43.6651 33.4911 34.4811 42.3598
0.2588 18.0 2268 0.9851 43.9449 34.4339 35.2144 42.7187
0.2588 19.0 2394 0.9968 43.1336 32.2189 33.169 41.8021
0.217 20.0 2520 1.0100 42.957 32.0188 32.6537 41.6299
0.217 21.0 2646 1.0294 43.5366 33.4319 34.3512 42.2255
0.1843 22.0 2772 1.0318 42.612 31.6127 32.6111 41.1739
0.1843 23.0 2898 1.0364 43.3555 33.0608 33.7772 42.0746
0.1621 24.0 3024 1.0472 43.551 33.1728 33.9075 42.3046
0.1621 25.0 3150 1.0489 43.3577 32.7069 33.2761 41.9641
0.1411 26.0 3276 1.0568 43.3813 32.9442 33.6935 42.014
0.1411 27.0 3402 1.0609 43.4543 33.1567 34.1049 42.0669
0.1313 28.0 3528 1.0629 43.5826 33.403 34.5062 42.2977
0.1313 29.0 3654 1.0672 43.4541 33.1488 34.242 42.1413
0.1196 30.0 3780 1.0661 43.385 32.8564 33.7117 42.0926

Framework versions

  • Transformers 4.45.1
  • Pytorch 2.4.1
  • Datasets 3.0.1
  • Tokenizers 0.20.0
Downloads last month
7
Safetensors
Model size
484M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for mirari/m2m100_418M-lecturaFacil

Finetuned
(80)
this model