---
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- unsloth
- generated_from_trainer
base_model: unsloth/tinyllama-bnb-4bit
datasets:
- generator
model-index:
- name: MunicipalPredictionModel-TinyLlama
  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. -->

# MunicipalPredictionModel-TinyLlama

This model is a fine-tuned version of [unsloth/tinyllama-bnb-4bit](https://huggingface.co/unsloth/tinyllama-bnb-4bit) on the generator dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0067

## 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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.4976        | 0.0431 | 10   | 1.3395          |
| 1.2954        | 0.0862 | 20   | 1.2420          |
| 1.2219        | 0.1293 | 30   | 1.1957          |
| 1.1914        | 0.1724 | 40   | 1.1660          |
| 1.1515        | 0.2155 | 50   | 1.1449          |
| 1.1491        | 0.2586 | 60   | 1.1301          |
| 1.129         | 0.3017 | 70   | 1.1172          |
| 1.1175        | 0.3448 | 80   | 1.1070          |
| 1.1129        | 0.3879 | 90   | 1.0982          |
| 1.088         | 0.4310 | 100  | 1.0902          |
| 1.0823        | 0.4741 | 110  | 1.0847          |
| 1.0912        | 0.5172 | 120  | 1.0789          |
| 1.0847        | 0.5603 | 130  | 1.0734          |
| 1.0773        | 0.6034 | 140  | 1.0689          |
| 1.0671        | 0.6466 | 150  | 1.0632          |
| 1.0711        | 0.6897 | 160  | 1.0602          |
| 1.0607        | 0.7328 | 170  | 1.0558          |
| 1.0579        | 0.7759 | 180  | 1.0515          |
| 1.0565        | 0.8190 | 190  | 1.0481          |
| 1.0605        | 0.8621 | 200  | 1.0459          |
| 1.0626        | 0.9052 | 210  | 1.0413          |
| 1.0363        | 0.9483 | 220  | 1.0392          |
| 1.0422        | 0.9914 | 230  | 1.0363          |
| 1.0365        | 1.0345 | 240  | 1.0359          |
| 1.0083        | 1.0776 | 250  | 1.0327          |
| 1.0138        | 1.1207 | 260  | 1.0297          |
| 1.015         | 1.1638 | 270  | 1.0275          |
| 1.0116        | 1.2069 | 280  | 1.0258          |
| 1.0168        | 1.25   | 290  | 1.0245          |
| 1.0147        | 1.2931 | 300  | 1.0225          |
| 1.022         | 1.3362 | 310  | 1.0203          |
| 1.0118        | 1.3793 | 320  | 1.0199          |
| 1.0069        | 1.4224 | 330  | 1.0181          |
| 1.0042        | 1.4655 | 340  | 1.0163          |
| 0.9963        | 1.5086 | 350  | 1.0145          |
| 1.0127        | 1.5517 | 360  | 1.0137          |
| 1.0076        | 1.5948 | 370  | 1.0125          |
| 1.0127        | 1.6379 | 380  | 1.0113          |
| 1.0048        | 1.6810 | 390  | 1.0102          |
| 0.9904        | 1.7241 | 400  | 1.0093          |
| 0.9971        | 1.7672 | 410  | 1.0091          |
| 1.0073        | 1.8103 | 420  | 1.0082          |
| 1.0138        | 1.8534 | 430  | 1.0077          |
| 0.9952        | 1.8966 | 440  | 1.0074          |
| 0.9977        | 1.9397 | 450  | 1.0070          |
| 0.9966        | 1.9828 | 460  | 1.0067          |


### Framework versions

- PEFT 0.12.0
- Transformers 4.43.4
- Pytorch 2.2.0+cu121
- Datasets 2.18.0
- Tokenizers 0.19.0