---
language:
  - fr
  - en
license: apache-2.0
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
  - merged
  - lora
  - peft
  - qwen2
  - cybersecurity
  - fine-tuned
model-index:
  - name: ISO27001-Expert-1.5B-Merged
    results: []
---

# ISO27001-Expert-1.5B — Merged (Standalone)

**Expert en norme ISO 27001 (Systemes de Management de la Securite de l'Information)**

This is the **merged / standalone** version of [AYI-NEDJIMI/ISO27001-Expert-1.5B](https://huggingface.co/AYI-NEDJIMI/ISO27001-Expert-1.5B).
The LoRA adapter weights have been fully merged into the base model ([Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)), so **no PEFT library is required** at inference time.

| Property | Value |
|---|---|
| **Base model** | [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) |
| **Adapter version** | [AYI-NEDJIMI/ISO27001-Expert-1.5B](https://huggingface.co/AYI-NEDJIMI/ISO27001-Expert-1.5B) |
| **Parameters** | 1.5B |
| **LoRA rank (r)** | 64 |
| **LoRA alpha** | 128 |
| **Precision** | float16 |
| **License** | Apache 2.0 |

## Description

- **FR** : Expert en norme ISO 27001 (Systemes de Management de la Securite de l'Information). Version fusionnee — chargement direct sans adaptateur PEFT.
- **EN** : ISO 27001 Information Security Management Systems expert. Merged version — direct loading without PEFT adapter.

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "AYI-NEDJIMI/ISO27001-Expert-1.5B-Merged"

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

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Explain the key principles of ISO 27001."},
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

> **Note:** No need to install or import `peft` — this model is fully standalone.

## Links

- **Adapter version (LoRA):** [AYI-NEDJIMI/ISO27001-Expert-1.5B](https://huggingface.co/AYI-NEDJIMI/ISO27001-Expert-1.5B)
- **Base model:** [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
- **Portfolio collection:** [AYI-NEDJIMI Collections](https://huggingface.co/collections/AYI-NEDJIMI)

## Training details

This model was fine-tuned using **LoRA** (Low-Rank Adaptation) with the following configuration:

- **LoRA rank (r):** 64
- **LoRA alpha:** 128
- **LoRA dropout:** 0.05
- **Target modules:** q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj

The adapter weights were then merged into the base model using `model.merge_and_unload()` from the PEFT library to produce this standalone checkpoint.