metadata
base_model: microsoft/Phi-3.5-mini-instruct
library_name: peft
license: mit
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: question-generator-v2
results: []
question-generator-v2
This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7497
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.0483 | 0.0967 | 50 | 0.9260 |
0.8577 | 0.1934 | 100 | 0.8202 |
0.7996 | 0.2901 | 150 | 0.7895 |
0.7802 | 0.3868 | 200 | 0.7784 |
0.7671 | 0.4836 | 250 | 0.7721 |
0.761 | 0.5803 | 300 | 0.7688 |
0.7587 | 0.6770 | 350 | 0.7663 |
0.7529 | 0.7737 | 400 | 0.7637 |
0.7562 | 0.8704 | 450 | 0.7616 |
0.7507 | 0.9671 | 500 | 0.7602 |
0.7274 | 1.0638 | 550 | 0.7589 |
0.7422 | 1.1605 | 600 | 0.7574 |
0.735 | 1.2573 | 650 | 0.7571 |
0.7367 | 1.3540 | 700 | 0.7555 |
0.7471 | 1.4507 | 750 | 0.7549 |
0.7404 | 1.5474 | 800 | 0.7541 |
0.742 | 1.6441 | 850 | 0.7533 |
0.7385 | 1.7408 | 900 | 0.7530 |
0.7352 | 1.8375 | 950 | 0.7525 |
0.7323 | 1.9342 | 1000 | 0.7516 |
0.7328 | 2.0309 | 1050 | 0.7515 |
0.7264 | 2.1277 | 1100 | 0.7510 |
0.704 | 2.2244 | 1150 | 0.7505 |
0.7242 | 2.3211 | 1200 | 0.7510 |
0.7203 | 2.4178 | 1250 | 0.7502 |
0.7285 | 2.5145 | 1300 | 0.7499 |
0.7192 | 2.6112 | 1350 | 0.7502 |
0.7204 | 2.7079 | 1400 | 0.7497 |
Framework versions
- PEFT 0.12.0
- Transformers 4.42.3
- Pytorch 2.1.2
- Datasets 2.20.0
- Tokenizers 0.19.1