Rebiha

Home › Datasets › Email & Professional Communication

Email & Professional Communication Fine-Tuning Dataset

Professional emails, replies, follow-ups, and workplace writing.

Fine-tune a model on this dataset

$19 for the dataset. Training is priced separately by model, starting at $47.

What is in this dataset

919 instruction-response pairs covering professional emails, replies, follow-ups, and workplace writing. Each pair is a realistic question with a full written answer, in the JSONL format used for supervised fine-tuning. No synthetic filler, no duplicated rows.

Example pairs

Real rows from the dataset, shown in full.

Write a cold outreach email to a senior marketing director at a potential partner company. I want to propose a collaboration on a joint webinar. The goal is to get a reply and spark interest.

Subject: Joint Webinar Idea: [Your Topic] Hi [First Name], I’ve been following [Company]’s recent work on [specific campaign or achievement], and I’m impressed by your approach to [specific area, e.g., B2B lead generation]. I’m reaching out because I think a joint webinar on [shared topic, e.g., ‘Scaling Content for Enterprise Audiences’] could deliver real value to both our audiences. At [Your Company], we specialize in [brief value proposition], and I believe combining our insights would create a compelling session. Would you be open to a quick 15-minute call next week to discuss potential formats and logistics? Let me know what works best for you. Best, [Your Name]

I sent a cold outreach email to a prospect five days ago and haven't heard back. Write a polite follow-up email that doesn't sound pushy but still encourages a reply.

Subject: Following Up: [Original Topic] Hi [First Name], I know you’re busy, so I wanted to briefly follow up on my earlier note about [specific offer, e.g., the joint webinar idea]. No pressure at all—just checking if this is something that might be on your radar. If timing isn’t right, a quick ‘not now’ is totally fine. I’d rather not clutter your inbox. Thanks for considering, and I hope you have a great week. Best, [Your Name]

I need to politely decline a meeting request from a colleague in another department. They want to discuss a project I’m not able to join due to capacity. Write a professional email that turns down the request but maintains goodwill.

Subject: Regarding Your Meeting Request Hi [Colleague’s Name], Thank you for the invitation to discuss [project name]. I appreciate you thinking of me. Unfortunately, due to my current workload and commitments, I won’t be able to join this meeting or participate in the project at this time. I don’t want to overcommit and deliver less than my best. I’d be happy to share any relevant notes or context in writing if that’s helpful. And please feel free to reach out again for future initiatives where my bandwidth allows. Best of luck with the project, [Your Name]

Who this is for

Writing tools that draft professional email at work.

Which models this works with

The dataset is plain JSONL, so it works with any instruction-tuned open model. On Rebiha you can train it directly on:

Training uses LoRA, or QLoRA on a 4-bit base for larger models. Your base model's weights are never modified — training produces a separate adapter.

How it works

  1. Pick this dataset and a base model.
  2. Rebiha runs the job on a managed GPU. You set up nothing.
  3. Download your model: a ready-to-run GGUF, plus developer assets (adapter, tokenizer, configs).

Fine-tuning or RAG?

Use RAG when the model needs facts that change — your prices, your documents, today's inventory. Use fine-tuning when you want consistent style and structure without prompting for it every time. This dataset teaches the shape of a good answer in this domain, not facts to look up.

Related datasets

Start fine-tuning