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_hom_100m_mult_reseg_ep20_goldfish

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

  • Loss: 4.8870

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: 1352
  • training_steps: 13525
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
7.1732 0.7394 500 6.2638
5.6194 1.4776 1000 5.4843
5.227 2.2159 1500 5.0583
4.841 2.9553 2000 4.7868
4.6153 3.6935 2500 4.6040
4.3649 4.4318 3000 4.4778
4.3025 5.1701 3500 4.3898
4.1416 5.9094 4000 4.3192
3.9741 6.6477 4500 4.2908
3.8149 7.3860 5000 4.2862
3.8348 8.1242 5500 4.2873
3.6977 8.8636 6000 4.2723
3.5328 9.6018 6500 4.3113
3.3815 10.3401 7000 4.3560
3.4349 11.0784 7500 4.3939
3.307 11.8177 8000 4.4125
3.1486 12.5560 8500 4.4781
3.0233 13.2943 9000 4.5435
3.0794 14.0325 9500 4.5834
2.9665 14.7719 10000 4.6196
2.8357 15.5102 10500 4.6898
2.7675 16.2484 11000 4.7474
2.7861 16.9878 11500 4.7618
2.7079 17.7261 12000 4.8156
2.6257 18.4643 12500 4.8569
2.6031 19.2026 13000 4.8812
2.5892 19.9420 13500 4.8869

Framework versions

  • Transformers 4.57.1
  • Pytorch 2.9.1+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