---
base_model: google/gemma-2-2b-jpn-it
language:
- multilingual
datasets:
  - mlabonne/orpo-dpo-mix-40k
library_name: transformers
license: gemma
license_link: https://ai.google.dev/gemma/terms
pipeline_tag: text-generation
tags:
- nlp
- code
quantized_by: ymcki
widget:
- messages:
  - role: user
    content: Can you provide ways to eat combinations of bananas and dragonfruits?
---

Original model: https://huggingface.co/google/gemma-2-2b-jpn-it

## Prompt format

```
<start_of_turn>user
{prompt}<end_of_turn>
<start_of_turn>model
<end_of_turn>
<start_of_turn>model

```

Note that this model does not support a System prompt.

This is abliterated model of [google/gemma-2-2b-jpn-it](https://huggingface.co/google/gemma-2-2b-jpn-it) using the 
[method](https://medium.com/@mlabonne/uncensor-any-llm-with-abliteration-d30148b7d43e) 
described by mlabonne.

Layer 17 of the original model was chosen for abliteration.
I also created another layer 18 and 24 abliterated model for comparison.

ORPO fine tuning was performed for four, eight and twelve epoches. Lowest eval
at the end of the fourth epoch was at 3.72 epoch. Lowest eval_loss at the 
end of the eighth epoch was 7.48 epoch. Lowest eval_loss at the end of the
twelve epoch was 11.96 epoch. Checkpoint at 11.96 epoch was chosen to generate this model.

| Epoch | loss | eval_loss | eval_logps/rejected | eval_logps/chosen |
| ----- | ---- | --------- | ------------------- | ----------------- |
| 1.00 | 1.2015 | 1.0501 | -1.0451 | -0.7449 |
| 2.00 | 1.2576 | 1.0145 | -1.1346 | -0.7248 |
| 3.00 | 0.9310 | 0.9958 | -1.2629 | -0.7332 |
| 3.72 | 0.7453 | 0.9848 | -1.2205 | -0.7006 |
| 4.00 | 0.8866 | 0.9857 | -1.2231 | -0.7019 |
| 5.00 | 0.8696 | 1.0204 | -1.2242 | -0.7523 |
| 6.00 | 0.9807 | 0.9959 | -1.3093 | -0.7257 |
| 7.00 | 0.3851 | 0.9687 | -1.3826 | -0.7103 |
| 7.48 | 1.2072 | 0.9638 | -1.4512 | -0.6959 |
| 8.00 | 1.4118 | 0.9653 | -1.5047 | -0.6990 |
| 9.00 | 1.1466 | 1.0070 | -1.6149 | -0.7567 |
| 10.00 | 1.4646 | 0.9801 | -1.9078 | -0.7207 |
| 11.00 | 1.8303 | 0.9620 | -2.0278 | -0.7096 |
| 11.96 | 0.9252 | 0.9372 | -2.0292 | -0.6692 |
| 12.00 | 1.1489 | 0.9560 | -1.9191 | -0.7226 |

The fine tuned model is uploaded here to be evaluated by the Open LLM Leaderboard to see if the slightly brain damaged non-ORPO model can be healed. Again, the fine tuning method is also based on one described by [mlabonne](https://towardsdatascience.com/fine-tune-llama-3-with-orpo-56cfab2f9ada) but the input model was read into VRAM by [unsloth](https://github.com/unslothai/unsloth) to allow using the full 40k dataset to run on a single 3090.

## Benchmark (100.0*raw scores only)

Click on the model name go to the raw score json generated by Open LLM Leaderboard.

| Model | Average | IFEval | BHH | Math Lv5 | GPQA | MUSR | MMLU-PRO |
| ----- | ------- | ------ | ----|--------- | ---- | ---- | -------- |
| [gemma-2-2b-jpn-it](https://huggingface.co/datasets/open-llm-leaderboard/results/blob/main/google/gemma-2-2b-jpn-it/results_2024-10-15T15-21-39.173019.json) | 30.82 | 54.11 | 41.43 | 0.0 | 27.52 | 37.17 | 24.67 |
| [gemma-2-2b-jpn-it-abliterated-17-ORPO (4 epoches)](https://huggingface.co/datasets/open-llm-leaderboard/results/raw/main/ymcki/gemma-2-2b-jpn-it-abliterated-17-ORPO/results_2024-10-20T02-46-59.069357.json) | 29.99 | 50.94 | 38.59 | 2.87 | 27.43 | 38.23 | 21.86 |
| [gemma-2-2b-jpn-it-abliterated-17-ORPO (8 epoches)](https://huggingface.co/datasets/open-llm-leaderboard/results/raw/main/ymcki/gemma-2-2b-jpn-it-abliterated-17-ORPO/results_2024-10-24T00-00-00.000000.json) | 29.42 | 48.95 | 38.27 | 3.17 | 26.93 | 37.43 | 21.77 |
| gemma-2-2b-jpn-it-abliterated-17-ORPO (12 epoches) | TBD | TBD | TBD | TBD | TBD | TBD | TBD |
| [gemma-2-2b-jpn-it-abliterated-18-ORPO (4 epoches)](https://huggingface.co/datasets/open-llm-leaderboard/results/raw/main/ymcki/gemma-2-2b-jpn-it-abliterated-18-ORPO/results_2024-10-22T04-04-56.385050.json) | 29.94 | 48.97 | 40.18 | 3.02 | 26.17 | 39.42 | 21.85 |
| [gemma-2-2b-jpn-it-abliterated-17](https://huggingface.co/datasets/open-llm-leaderboard/results/raw/main/ymcki/gemma-2-2b-jpn-it-abliterated-17/results_2024-10-18T15-18-46.821674.json) | 30.29 | 52.65 | 40.46 | 0.0 | 27.18 | 36.90 | 24.55 |
| [gemma-2-2b-jpn-it-abliterated-18](https://huggingface.co/datasets/open-llm-leaderboard/results/raw/main/ymcki/gemma-2-2b-jpn-it-abliterated-18/results_2024-10-18T15-41-42.399571.json) | 30.61 | 53.02 | 40.96 | 0.0 | 27.35 | 37.30 | 25.05 |
| [gemma-2-2b-jpn-it-abliterated-24](https://huggingface.co/datasets/open-llm-leaderboard/results/raw/main/ymcki/gemma-2-2b-jpn-it-abliterated-24/results_2024-10-25T16-29-46.542899.json) | 30.61 | 51.37 | 40.77 | 0.0 | 27.77 | 39.02 | 24.73 |

Looks like fine tuning for 8 epoches is still not enough. May need to run more epoches.

## How to run this model

```py
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch

model_id = "gemma-2-2b-jpn-it-abliterated-17-ORPO"
dtype = torch.bfloat16

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="cuda",
    torch_dtype=dtype,)

chat = [
    { "role": "user", "content": "Write a hello world program" },
]
prompt = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
```

## Downloading using huggingface-cli

First, make sure you have hugginface-cli installed:

```
pip install -U "huggingface_hub[cli]"
```

Then, you can target the specific file you want:

```
huggingface-cli download ymcki/gemma-2-2b-jpn-it-abliterated-17-ORPO --include "*" --local-dir ./
```

## Credits

Thank you mlabonne for describing his fine tuning method.

Thanks FullOf_Bad_Ideas from LocalLlama for the suggestion of using unsloth to save VRAM.