---
library_name: peft
license: other
base_model: microsoft/Phi-3.5-mini-instruct
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: cereborn-de-3step
  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. -->

# cereborn-de-3step

This model is a fine-tuned version of [thomheinrich/CEREBORN-german](https://huggingface.co/thomheinrich/CEREBORN-german) on the 3step dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2900
- Num Input Tokens Seen: 630048

## 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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- num_epochs: 10.0

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Input Tokens Seen |
|:-------------:|:------:|:----:|:---------------:|:-----------------:|
| 0.6911        | 7.1290 | 50   | 1.2630          | 451904            |


### Framework versions

- PEFT 0.12.0
- Transformers 4.49.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0