---
license: apache-2.0
datasets:
- Intel/orca_dpo_pairs
tags:
- mistral
- dpo
- una
- finetune
- chatml
- instruct
---
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6468ce47e134d050a58aa89c/-BlRd-74Hk4B153wl_ryD.png)

# Neural-una-cybertron-7b

Neural-una-cybertron-7b is an [fblgit/una-cybertron-7b-v2-bf16](https://huggingface.co/fblgit/una-cybertron-7b-v2-bf16) model that has been further fine-tuned with Direct Preference Optimization (DPO) using the [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs) dataset.

This model was created after examining the procedure of [mlabonne/NeuralHermes-2.5-Mistral-7B](https://hf.co/mlabonne/NeuralHermes-2.5-Mistral-7B) model. Special thanks to [@mlabonne](https://hf.co/mlabonne).

## Addionatal Information 

This model was fine-tuned on `Nvidia A100-SXM4-40GB` GPU.

The total training time was 1 hour and 10 minutes.

# Prompt Template(s)

### ChatML

```
<|im_start|>system
{system}<|im_end|>
<|im_start|>user
{user}<|im_end|>
<|im_start|>assistant
{asistant}<|im_end|>
```

## Training hyperparameters

**LoRA**:
* r=16
* lora_alpha=16
* lora_dropout=0.05
* bias="none"
* task_type="CAUSAL_LM"
* target_modules=['k_proj', 'gate_proj', 'v_proj', 'up_proj', 'q_proj', 'o_proj', 'down_proj']

**Training arguments**:
* per_device_train_batch_size=4
* gradient_accumulation_steps=4
* gradient_checkpointing=True
* learning_rate=5e-5
* lr_scheduler_type="cosine"
* max_steps=200
* optim="paged_adamw_32bit"
* warmup_steps=100

**DPOTrainer**:
* beta=0.1
* max_prompt_length=1024
* max_length=1536