---
license: gemma
base_model: google/gemma-7b
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
- trl
- sft
- generated_from_trainer
datasets:
- masakhane/african-ultrachat
model-index:
- name: zephyr-7b-gemma-sft-african-ultrachat
  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. -->

# zephyr-7b-gemma-sft-african-ultrachat

This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the masakhane/african-ultrachat dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0802

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

[masakhane/african-ultrachat](https://huggingface.co/datasets/masakhane/african-ultrachat)

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.1942        | 1.0   | 2089 | 1.1757          |
| 0.952         | 2.0   | 4178 | 1.0642          |
| 0.7033        | 3.0   | 6267 | 1.0802          |


### Framework versions

- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.14.6
- Tokenizers 0.15.2


### How to use 

``` python
import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="masakhane/zephyr-7b-gemma-sft-african-ultrachat", torch_dtype=torch.bfloat16, device_map="auto")

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
    {
        "role": "system",
        "content": "You are a friendly chatbot who always responds in the style of a pirate",
    },
    {"role": "user", "content": "ሰላም እንዴት ነህ?"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])


# <|system|>
# You are a friendly chatbot who always responds in the style of a pirate<eos>
# <|user|>
# ሰላም እንዴት ነህ?<eos>
# <|assistant|>
# ሰላም እንዴት ነህ/ነሽ? እኔም በጤና ነኝ። እንደምን ነህ/ነሽ እና የምትፈልገው እንዴት ነው?
```