Rebiha

Home › Datasets › Language Learning

Language Learning Fine-Tuning Dataset

Language learning, grammar, vocabulary, pronunciation

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

955 instruction-response pairs covering language learning, grammar, vocabulary, pronunciation. 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.

Explain the difference between 'ser' and 'estar' in Spanish, with examples.

In Spanish, both 'ser' and 'estar' mean 'to be,' but they’re used in distinct contexts. Use 'ser' for permanent or inherent characteristics—identity, origin, time, and essential qualities. For example: 'Ella es médica' (She is a doctor, a profession as identity) and 'Hoy es lunes' (Today is Monday). Use 'estar' for temporary states, locations, and conditions—emotions, positions, and ongoing actions. For example: 'Estoy cansado' (I am tired, a temporary state) and 'El libro está en la mesa' (The book is on the table, a location). A key trick: 'ser' for what something is; 'estar' for…

Teach me five essential phrases for ordering food in a French restaurant.

Here are five practical French phrases for ordering food: 1) 'Je voudrais…' (I would like…) – use for ordering, e.g., 'Je voudrais un croissant.' 2) 'L'addition, s'il vous plaît' (The check, please) – request the bill politely. 3) 'Qu'est-ce que vous recommandez ?' (What do you recommend?) – ask for suggestions from the server. 4) 'Sans gluten' (Gluten-free) – specify dietary needs. 5) 'Je suis allergique aux…' (I am allergic to…) – state allergies, e.g., 'Je suis allergique aux arachides' (I am allergic to peanuts). Practice these with the polite 'vous' form to sound courteous.

Explain how to conjugate regular -ar verbs in the present tense in Spanish, using 'hablar' as an example.

To conjugate regular -ar verbs in Spanish present tense, remove the infinitive ending '-ar' and add the following endings: -o, -as, -a, -amos, -áis, -an. Using 'hablar' (to speak): 'yo hablo' (I speak), 'tú hablas' (you speak, informal), 'él/ella/usted habla' (he/she/you formal speak), 'nosotros hablamos' (we speak), 'vosotros habláis' (you all speak, Spain), 'ellos/ellas/ustedes hablan' (they/you all speak). Notice the accent on the 'á' in 'habláis' to maintain stress. This pattern works for all regular -ar verbs like 'cantar' (to sing) or 'estudiar' (to study).

Who this is for

Language-learning and translation tools covering grammar and pronunciation.

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