You need to agree to share your contact information to access this model

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

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

dense_eng_100m_mult_het_retok-het

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

  • Loss: 4.8737

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: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 723
  • training_steps: 7235
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
8.419 0.6908 500 7.7970
6.2543 1.3813 1000 6.0385
5.6668 2.0718 1500 5.4910
5.2471 2.7627 2000 5.2379
5.0071 3.4532 2500 5.0902
4.834 4.1437 3000 4.9959
4.7398 4.8345 3500 4.9203
4.5632 5.5250 4000 4.8889
4.5025 6.2155 4500 4.8730
4.404 6.9064 5000 4.8471
4.2675 7.5969 5500 4.8560
4.1516 8.2874 6000 4.8683
4.1668 8.9782 6500 4.8611
4.0815 9.6687 7000 4.8746

Framework versions

  • Transformers 4.57.1
  • Pytorch 2.9.0+cu128
  • Datasets 3.6.0
  • Tokenizers 0.22.1
Downloads last month
-
Safetensors
Model size
0.2B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support