Edit model card

V0508HMA15HPHI3B2

This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0885

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: 0.0003
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 16
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
2.3691 0.09 10 0.1863
0.1586 0.18 20 0.1406
0.1444 0.27 30 0.1391
0.1366 0.36 40 0.1237
0.1228 0.45 50 0.1353
0.1212 0.54 60 0.0898
0.1116 0.63 70 0.1008
0.0983 0.73 80 0.0803
0.0756 0.82 90 0.0930
0.0848 0.91 100 0.0721
0.073 1.0 110 0.0729
0.0499 1.09 120 0.0691
0.0594 1.18 130 0.1105
0.067 1.27 140 0.0751
0.0489 1.36 150 0.0821
0.0622 1.45 160 0.0838
0.0654 1.54 170 0.0764
0.0574 1.63 180 0.0826
0.0562 1.72 190 0.0757
0.0608 1.81 200 0.0795
0.061 1.9 210 0.0796
0.0552 1.99 220 0.0796
0.0293 2.08 230 0.0849
0.0219 2.18 240 0.1143
0.0257 2.27 250 0.0967
0.0204 2.36 260 0.0831
0.0251 2.45 270 0.0882
0.0182 2.54 280 0.0959
0.0189 2.63 290 0.0925
0.0243 2.72 300 0.0909
0.0226 2.81 310 0.0890
0.017 2.9 320 0.0884
0.0201 2.99 330 0.0885

Framework versions

  • Transformers 4.36.0.dev0
  • Pytorch 2.1.2+cu121
  • Datasets 2.18.0
  • Tokenizers 0.14.1
Downloads last month

-

Downloads are not tracked for this model. How to track
Safetensors
Model size
3.9B params
Tensor type
F32
·
BF16
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for Litzy619/V0508HMA15HPHI3B2

Finetuned
this model