---
base_model: microsoft/Phi-3.5-mini-instruct
library_name: peft
license: mit
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: Phi-3.5-MultiCap-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. -->

# Phi-3.5-MultiCap-3

This model is a fine-tuned version of [microsoft/Phi-3.5-mini-instruct](https://huggingface.co/microsoft/Phi-3.5-mini-instruct) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5216

## 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
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.8409        | 0.2256 | 50   | 0.8368          |
| 0.6072        | 0.4512 | 100  | 0.6126          |
| 0.5646        | 0.6768 | 150  | 0.5713          |
| 0.611         | 0.9024 | 200  | 0.5536          |
| 0.5388        | 1.1280 | 250  | 0.5436          |
| 0.5283        | 1.3536 | 300  | 0.5369          |
| 0.4997        | 1.5792 | 350  | 0.5324          |
| 0.5482        | 1.8049 | 400  | 0.5287          |
| 0.535         | 2.0305 | 450  | 0.5262          |
| 0.4813        | 2.2561 | 500  | 0.5243          |
| 0.5266        | 2.4817 | 550  | 0.5229          |
| 0.5341        | 2.7073 | 600  | 0.5221          |
| 0.5447        | 2.9329 | 650  | 0.5216          |


### Framework versions

- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1