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.

gpt2moe_het2_100mb

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

  • Loss: 4.2740

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: 4
  • 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: 7404
  • training_steps: 74047
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
No log 0 0 11.0993
7.3615 0.2701 2000 7.0780
6.47 0.5402 4000 6.1761
5.9733 0.8103 6000 5.6818
5.6141 1.0804 8000 5.3557
5.4025 1.3504 10000 5.1035
5.2257 1.6205 12000 4.9233
5.1025 1.8906 14000 4.8059
4.9549 2.1607 16000 4.7256
4.9029 2.4308 18000 4.6622
4.8632 2.7009 20000 4.6041
4.8398 2.9710 22000 4.5621
4.8182 3.0 22215 4.5559
4.6986 3.2411 24000 4.5316
4.6922 3.5111 26000 4.5008
4.6772 3.7812 28000 4.4704
4.5482 4.0513 30000 4.4513
4.557 4.3214 32000 4.4324
4.5699 4.5915 34000 4.4124
4.5508 4.8616 36000 4.3927
4.4423 5.1317 38000 4.3846
4.4491 5.4018 40000 4.3701
4.4602 5.6718 42000 4.3575
4.4429 5.9419 44000 4.3404
4.353 6.2120 46000 4.3403
4.3662 6.4821 48000 4.3306
4.3708 6.7522 50000 4.3197
4.29 7.0223 52000 4.3150
4.2882 7.2924 54000 4.3123
4.2945 7.5625 56000 4.3045
4.3034 7.8325 58000 4.2956
4.2248 8.1026 60000 4.2956
4.2257 8.3727 62000 4.2925
4.2318 8.6428 64000 4.2852
4.236 8.9129 66000 4.2798
4.1746 9.1830 68000 4.2823
4.1798 9.4531 70000 4.2792
4.1827 9.7232 72000 4.2759
4.1743 9.9932 74000 4.2740

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