Access request

This repository is publicly accessible, but you must accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

dyu-nllb-600M-fr2dyu

This model is a fine-tuned version of goaicorp/dyu-nllb-600M-fr2dyu on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 3.1010
  • Bleu: 5.3196
  • Gen Len: 67.0659

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: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • 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: cosine
  • lr_scheduler_warmup_steps: 700
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
No log 1.0 71 5.5247 1.9734 69.7723
24.7282 2.0 142 4.3075 2.3248 65.3766
19.2418 3.0 213 3.8885 2.9509 13.261
19.2418 4.0 284 3.6824 3.6019 13.1027
17.0945 5.0 355 3.5412 4.2403 13.2957
15.9574 6.0 426 3.4223 4.5538 13.3249
15.9574 7.0 497 3.2880 5.068 67.5982
15.0356 8.0 568 3.2063 5.1631 68.4521
13.9984 9.0 639 3.1440 5.638 13.2277
13.1780 10.0 710 3.1010 5.3196 67.0659

Framework versions

  • Transformers 5.5.0
  • Pytorch 2.8.0+cu128
  • Datasets 4.8.4
  • Tokenizers 0.22.2
Downloads last month
8
Safetensors
Model size
0.6B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for goaicorp/dyu-nllb-600M-fr2dyu

Unable to build the model tree, the base model loops to the model itself. Learn more.