Hugging Face's logo

Gunslinger3D
/
fine-tuning-Phi2-with-webglm-qa-with-lora_4

PEFT
Safetensors
Generated from Trainer
Model card Files Community
fine-tuning-Phi2-with-webglm-qa-with-lora_4 / README.md
Gunslinger3D's picture
Gunslinger3D
fine-tuning-Phi2-with-webglm-qa-with-lora_4
b83c80a verified
preview code
|
raw
history blame
3.99 kB
metadata
license: mit
library_name: peft
tags:
  - generated_from_trainer
base_model: microsoft/phi-2
model-index:
  - name: fine-tuning-Phi2-with-webglm-qa-with-lora_4
    results: []

fine-tuning-Phi2-with-webglm-qa-with-lora_4

This model is a fine-tuned version of microsoft/phi-2 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2392

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: 5e-05
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 5
  • total_train_batch_size: 10
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • training_steps: 500
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
8.19 0.2 10 7.9966
8.0261 0.4 20 7.7896
7.3527 0.6 30 7.2580
6.9568 0.8 40 5.9952
5.2411 1.0 50 3.7880
2.9772 1.2 60 1.8751
1.2384 1.39 70 0.7517
0.6916 1.59 80 0.6684
0.5669 1.79 90 0.6138
0.5195 1.99 100 0.5846
0.5281 2.19 110 0.5607
0.4764 2.39 120 0.5396
0.4655 2.59 130 0.5190
0.4787 2.79 140 0.4980
0.427 2.99 150 0.4765
0.41 3.19 160 0.4547
0.397 3.39 170 0.4317
0.3648 3.59 180 0.4087
0.3436 3.78 190 0.3863
0.3415 3.98 200 0.3661
0.3072 4.18 210 0.3481
0.2681 4.38 220 0.3341
0.3068 4.58 230 0.3201
0.2526 4.78 240 0.3095
0.2632 4.98 250 0.3003
0.2693 5.18 260 0.2936
0.2194 5.38 270 0.2874
0.2474 5.58 280 0.2826
0.2467 5.78 290 0.2770
0.2188 5.98 300 0.2726
0.2305 6.18 310 0.2690
0.2336 6.37 320 0.2643
0.2192 6.57 330 0.2614
0.2189 6.77 340 0.2588
0.2049 6.97 350 0.2564
0.2096 7.17 360 0.2540
0.221 7.37 370 0.2521
0.2167 7.57 380 0.2498
0.203 7.77 390 0.2484
0.1999 7.97 400 0.2469
0.1888 8.17 410 0.2458
0.195 8.37 420 0.2443
0.2358 8.57 430 0.2429
0.1929 8.76 440 0.2419
0.2066 8.96 450 0.2412
0.2101 9.16 460 0.2407
0.2009 9.36 470 0.2400
0.1976 9.56 480 0.2394
0.2013 9.76 490 0.2392
0.1956 9.96 500 0.2392

Framework versions

  • PEFT 0.7.1
  • Transformers 4.36.2
  • Pytorch 2.0.0
  • Datasets 2.15.0
  • Tokenizers 0.15.0