File size: 3,009 Bytes
2ec4e74 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
---
base_model: microsoft/Phi-3-mini-4k-instruct
library_name: peft
license: mit
tags:
- trl
- orpo
- generated_from_trainer
model-index:
- name: ORPO-PHI-3
results: []
---
<!-- 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. -->
# ORPO-PHI-3
This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.7712
- Rewards/chosen: -0.1577
- Rewards/rejected: -0.1527
- Rewards/accuracies: 0.3000
- Rewards/margins: -0.0050
- Logps/rejected: -1.5273
- Logps/chosen: -1.5771
- Logits/rejected: 2.7883
- Logits/chosen: 1.8098
- Nll Loss: 1.6979
- Log Odds Ratio: -0.7331
- Log Odds Chosen: -0.0576
## 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: 8e-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 1
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen | Nll Loss | Log Odds Ratio | Log Odds Chosen |
|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|:--------:|:--------------:|:---------------:|
| 1.7534 | 0.2020 | 25 | 1.7712 | -0.1577 | -0.1527 | 0.3000 | -0.0050 | -1.5273 | -1.5771 | 2.7883 | 1.8098 | 1.6979 | -0.7331 | -0.0576 |
| 1.9166 | 0.4040 | 50 | 1.7712 | -0.1577 | -0.1527 | 0.3000 | -0.0050 | -1.5273 | -1.5771 | 2.7883 | 1.8098 | 1.6979 | -0.7331 | -0.0576 |
| 1.436 | 0.6061 | 75 | 1.7712 | -0.1577 | -0.1527 | 0.3000 | -0.0050 | -1.5273 | -1.5771 | 2.7883 | 1.8098 | 1.6979 | -0.7331 | -0.0576 |
| 1.6618 | 0.8081 | 100 | 1.7712 | -0.1577 | -0.1527 | 0.3000 | -0.0050 | -1.5273 | -1.5771 | 2.7883 | 1.8098 | 1.6979 | -0.7331 | -0.0576 |
### Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1 |