metadata
license: mit
library_name: peft
tags:
- generated_from_trainer
base_model: microsoft/phi-2
model-index:
- name: final-checkpoint
results: []
final-checkpoint
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: 1.3170
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: 0.0002
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- training_steps: 1000
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.6634 | 0.05 | 25 | 1.3938 |
1.1888 | 0.1 | 50 | 1.3917 |
1.4465 | 0.15 | 75 | 1.3522 |
1.205 | 0.2 | 100 | 1.3718 |
1.4403 | 0.25 | 125 | 1.3429 |
1.1372 | 0.3 | 150 | 1.3639 |
1.4064 | 0.35 | 175 | 1.3408 |
1.1466 | 0.4 | 200 | 1.3473 |
1.4435 | 0.45 | 225 | 1.3355 |
1.2273 | 0.5 | 250 | 1.3414 |
1.4578 | 0.55 | 275 | 1.3364 |
1.1637 | 0.6 | 300 | 1.3362 |
1.429 | 0.65 | 325 | 1.3305 |
1.2027 | 0.7 | 350 | 1.3306 |
1.3971 | 0.75 | 375 | 1.3287 |
1.184 | 0.8 | 400 | 1.3312 |
1.4383 | 0.85 | 425 | 1.3262 |
1.1996 | 0.9 | 450 | 1.3271 |
1.448 | 0.95 | 475 | 1.3250 |
1.1902 | 1.0 | 500 | 1.3277 |
1.4134 | 1.05 | 525 | 1.3237 |
1.1459 | 1.1 | 550 | 1.3263 |
1.3473 | 1.15 | 575 | 1.3225 |
1.1879 | 1.2 | 600 | 1.3232 |
1.3742 | 1.25 | 625 | 1.3217 |
1.1129 | 1.3 | 650 | 1.3225 |
1.3756 | 1.35 | 675 | 1.3201 |
1.1268 | 1.4 | 700 | 1.3216 |
1.4186 | 1.45 | 725 | 1.3200 |
1.1602 | 1.5 | 750 | 1.3196 |
1.4368 | 1.55 | 775 | 1.3189 |
1.1397 | 1.6 | 800 | 1.3193 |
1.3991 | 1.65 | 825 | 1.3186 |
1.1293 | 1.7 | 850 | 1.3190 |
1.4509 | 1.75 | 875 | 1.3177 |
1.1433 | 1.8 | 900 | 1.3175 |
1.3638 | 1.85 | 925 | 1.3170 |
1.144 | 1.9 | 950 | 1.3172 |
1.3376 | 1.95 | 975 | 1.3170 |
1.1152 | 2.0 | 1000 | 1.3170 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.3
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.0