CoolFace
Modelpublic

edumunozsala/phi-3-mini-QLoRA

sourceHugging Facemitupdated 2y agoView on Hugging Face
0likes124downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

phi-3-mini-QLoRA

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: 0.5741

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
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —gradientaccumulationsteps: 8
  • —totaltrainbatch_size: 32
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_ratio: 0.1
  • —num_epochs: 3

Training results

Training LossEpochStepValidation Loss
1.14220.18101000.6625
0.62380.36192000.6002
0.59320.54293000.5906
0.59260.72384000.5860
0.57940.90485000.5834
0.58681.08576000.5815
0.57111.26677000.5796
0.57291.44768000.5785
0.58581.62869000.5772
0.57321.809510000.5763
0.57361.990511000.5756
0.56382.171512000.5754
0.57692.352413000.5746
0.56682.533414000.5745
0.56752.714315000.5742
0.56932.895316000.5741

Framework versions

  • —PEFT 0.10.0
  • —Transformers 4.40.2
  • —Pytorch 2.2.1+cu121
  • —Datasets 2.19.1
  • —Tokenizers 0.19.1