electron-rare/mascarade-embedded-dataset

Mascarade — Embedded Systems Q&A

Description

Q&A bilingue (FR/EN) sur les systèmes embarqués génériques : RTOS (FreeRTOS, Zephyr), Linux embarqué (Yocto, Buildroot), device tree, bootloaders (U-Boot, BL31), drivers noyau, et toolchains croisées.

Ce dataset fait partie de la famille Mascarade, un corpus thématique destiné au fine-tuning LoRA de modèles compacts (cible : Gemma-3n-E4B et équivalents) pour des assistants spécialisés en électronique embarquée.

Format : JSONL « ShareGPT-style » avec un tour system (rôle d'expert), un tour human (question) et un tour gpt (réponse complète, code compilable, schémas ASCII si pertinent).

Data sources (EU AI Act Template — AI Office, July 2025)

Publicly available datasets

None.

Web scraping

• ~10–15% scraped from [Stack Exchange Electronics] (CC-BY-SA-4.0) and [EEVblog Forum] (public domain) — per-sample URL + author attribution incomplete for the Stack Exchange portion (legacy collection, see warning on the mascarade-power/-dsp/-emc/-kicad companions).

Synthetically generated

• ~85% generated by LLM (RTOS, device tree, drivers, bootloader scenarios).

Licensed data

None.

Sample format

{
  "conversations": [
    {
      "from": "system",
      "value": "You are an expert embedded systems engineer specializing in ARM Cortex-M (STM32, nRF52, Teensy, SAMD), ESP32/ESP-IDF, and RISC-V architectures. You write production-grade firmware in C/C++, bare-metal and RTOS-based, with proper register-level access, DMA, interrupt handling, low-power modes, and bootloader design. You know CMSIS, HAL, LL drivers, linker scripts, startup code, and ARM/RISC-V assembly. You also handle Raspberry Pi bare-metal and embedded Linux (device trees, kernel modules, GPIO/SPI/I2C)."
    },
    {
      "from": "human",
      "value": "Write bare-metal ARM Cortex-M4 startup code in assembly with vector table and Reset_Handler."
    },
    {
      "from": "gpt",
      "value": "ARM Cortex-M4 bare-metal startup (GNU AS syntax):\n\n```asm\n/* startup.s — Cortex-M4 startup for STM32F4xx */\n.syntax unified\n.cpu cortex-m4\n.fpu fpv4-sp-d16\n.thumb\n\n/* Vector table — must be at 0x00000000 (or VTOR offset) */\n.section .isr_vector, \"a\", %progbits\n.type g_pfnVectors, %object\n\ng_pfnVectors:\n    .word _estack              /* Initial SP (top of RAM) */\n    .word Reset_Handler        /* Reset */\n    .word NMI_Handler          /* NMI */\n    .word HardFault_Handler    /* Hard Fault */\n    .word MemManage_Handler    /* MPU Fault */\n    .word BusFault_Handler     /* Bus Fault */\n    .word UsageFault_Handler   /* Usage Fault */\n    .word 0, 0, 0, 0           /* Reserved */\n    .word SVC_Handler          /* SVCall */\n    .word DebugMon_Handler     /* Debug Monitor */\n    .word 0     ..."
    }
  ]
}

Statistics

Métrique	Valeur
Total samples	8,344
Size	16.72 MB
Format	jsonl (ShareGPT conversations)
Languages	French / English mix
File	embedded_chat.jsonl

Usage

from datasets import load_dataset

ds = load_dataset("electron-rare/mascarade-embedded-dataset")
print(ds["train"][0]["conversations"])

Pour un fine-tuning ShareGPT-style direct (axolotl, unsloth, mlx-lm) :

# axolotl config
datasets:
  - path: electron-rare/mascarade-embedded-dataset
    type: sharegpt
    conversation: chatml

Licenses applied

This aggregated dataset is released under CC-BY-SA-4.0. Per-sample original licenses preserved in metadata.license when known.

Copyright considerations

• Stack Exchange content: CC-BY-SA-4.0 (compatible upgrade ; full attribution remediation in progress for the ~10–15% scraped portion).
• EEVblog content: public domain per forum rules.
• Synthetic LLM outputs: belong to dataset author per OpenAI/Anthropic ToS.

Opt-out: contact c.saillant@gmail.com. We respect TDMRep, robots.txt, and noai/noimageai signals. We honor Article 4(3) DSM Directive opt-outs.

License & EU AI Act

CC-BY-SA-4.0 (attribution + sharealike).

Données collectées et générées dans le cadre du projet electron-rare (fine-tuning LoRA sur Gemma-3n-E4B pour applications électronique embarquée).

Compatible EU AI Act : voir les signataires du GPAI Code of Practice (Anthropic, Mistral, Google) et la documentation transparence du projet : electron-bench.

Audit log: see docs/audit_kicad9plus.md for the legal-attribution audit performed on 2026-05-11.

Citation

@dataset{electron_rare_embedded_2026,
  author    = {electron-rare},
  title     = { Mascarade — Embedded Systems Q&A },
  year      = {2026},
  publisher = {Hugging Face},
  license   = {CC-BY-SA-4.0},
  url       = {https://huggingface.co/datasets/electron-rare/mascarade-embedded-dataset}
}

Related datasets

Famille electron-rare/mascarade-* couvrant : STM32, SPICE, KiCad, IoT, Power, DSP, EMC, embedded.

• mascarade-stm32-dataset
• mascarade-spice-dataset
• mascarade-kicad-dataset
• mascarade-iot-dataset
• mascarade-power-dataset
• mascarade-dsp-dataset
• mascarade-emc-dataset
• mascarade-embedded-dataset
• kill-life-embedded-qa — Q&A spécifique au projet Kill_LIFE
• kicad9plus-sch-corpus — corpus de schémas KiCad 9+