DuoNeural
ml-ai-engineer-sft
DuoNeural ML/AI Engineer SFT Dataset
A synthetic instruction-tuning dataset for training an LLM to be a useful pairing partner on ML/AI engineering work — debugging training runs, reasoning about architecture and infra choices, reviewing experiment design, and explaining core ML concepts with the specificity of someone who's actually run the experiments.
Why this dataset exists
Most general instruction-tuning data treats ML engineering questions the same as any… See the full description on the dataset page: https://huggingface.co/datasets/DuoNeural/ml-ai-engineer-sft.Gemma4-E2B-SFT-WebCode
Gemma4-E2B-SFT-WebCode
Synthetic frontend web development dataset. Natural language component description → production-ready code.
Frameworks: React, TypeScript, Tailwind CSS, Vanilla HTML/CSS/JS.
Components: Navigation, forms, modals, data tables, charts, infinite scroll, etc.
Format: ShareGPT/ChatML. Includes accessibility attributes and comments.
Use: Fine-tune models for frontend copilot tasks.
Generator: DuoNeural/TurboGemma4E2B, temperature 0.65.
cot-reasoning-2k
DuoNeural CoT Reasoning Dataset (2K)
A compact, high-quality chain-of-thought reasoning dataset generated for supervised fine-tuning (SFT). All 2,151 examples are quality-scored 5/5 and focus on explicit step-by-step reasoning traces.
Benchmark Results
Fine-tuned Qwen2.5-1.5B-Instruct on this dataset (3 epochs, LoRA rank 16, ~36 min on RTX 3090):
Metric
Baseline
Post-SFT
Δ Absolute
Δ Relative
GSM8K (flexible-extract)
0.3177
0.4890
+17.1pp
+53.9%
GSM8K… See the full description on the dataset page: https://huggingface.co/datasets/DuoNeural/cot-reasoning-2k.smollm2-think-dataset-run1Gemma4-E2B-SFT-SQL
Gemma4-E2B-SFT-SQL
Synthetic text-to-SQL dataset covering real-world database schemas.
Schemas: E-commerce, healthcare, SaaS analytics.
Query types: Joins, subqueries, aggregations, window functions, CTEs.
Format: ShareGPT/ChatML. Natural language question + SQL answer + brief explanation.
Use: Fine-tune models for autonomous database querying and agentic SQL generation.
Generator: DuoNeural/TurboGemma4E2B, temperature 0.4.
Gemma4-E2B-SFT-CoT
Gemma4-E2B-SFT-CoT
Synthetic chain-of-thought reasoning dataset generated by DuoNeural/TurboGemma4E2B (Gemma 4 E2B abliterated).
Generation: 2-pass synthesis + self-evaluation (FineWeb-Edu style), only examples scoring ≥4/5 retained.
Topics: Math, logic, physics, probability, algorithm analysis, number theory.
Format: ShareGPT/ChatML (messages column, user+assistant turns).
Use: Fine-tune small models for step-by-step reasoning capabilities.
Generator: DuoNeural/TurboGemma4E2B with… See the full description on the dataset page: https://huggingface.co/datasets/DuoNeural/Gemma4-E2B-SFT-CoT.