akkadian_translation

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 6.5583

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: 16
  • eval_batch_size: 16
  • seed: 42
  • 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: linear
  • num_epochs: 20
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
51.6427 1.0 88 43.5929
35.4809 2.0 176 29.6570
26.2054 3.0 264 23.5765
23.6990 4.0 352 19.5528
20.2700 5.0 440 16.8286
17.4142 6.0 528 15.0303
16.0694 7.0 616 13.5486
14.6152 8.0 704 12.1864
13.2197 9.0 792 10.8778
12.6036 10.0 880 10.0926
11.6968 11.0 968 9.4187
10.7210 12.0 1056 8.8837
10.1111 13.0 1144 8.2965
10.0251 14.0 1232 7.8286
8.8541 15.0 1320 7.3692
9.1151 16.0 1408 7.0018
8.4213 17.0 1496 6.7866
8.3269 18.0 1584 6.6407
8.4181 19.0 1672 6.5861
8.7340 20.0 1760 6.5583

Framework versions

  • Transformers 5.2.0
  • Pytorch 2.9.0+cu126
  • Datasets 4.0.0
  • Tokenizers 0.22.2
Downloads last month
3
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 saivineetha/akkadian_translation

Base model

google/mt5-small
Finetuned
(666)
this model