Update README.md

README.md

@@ -1,21 +1,118 @@
----
-base_model: unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit
-tags:
-- text-generation-inference
-- transformers
-- unsloth
-- qwen3
-license: apache-2.0
-language:
-- en
----
-
-# Uploaded finetuned  model
-
-- **Developed by:** MikaLabs
-- **License:** apache-2.0
-- **Finetuned from model :** unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit
-
-This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
-
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
+---
+license: apache-2.0
+base_model: unsloth/Qwen3-4B-Instruct-2507
+tags:
+  - education
+  - teaching
+  - worksheet-generation
+  - lesson-planning
+language:
+  - en
+pipeline_tag: text-generation
+---
+
+# Vector-L1-4B
+
+**Vector-L1-4B** is an open language model built by **MikaLabs** to help teachers create classroom materials — differentiated worksheets, lesson plans, quizzes, mark schemes, misconception guides, and tailored explanations across Maths and the Sciences.
+
+The "L1" denotes **Light, version 1** — the first and smallest member of a planned Vector model family. It is designed to run on modest consumer hardware so that schools and individual teachers can use it locally and offline.
+
+---
+
+## Model Summary
+
+| | |
+|---|---|
+| **Developed by** | MikaLabs |
+| **Model name** | Vector-L1-4B |
+| **License** | Apache 2.0 |
+| **Language** | English |
+| **Domain** | K–12 / secondary education: Maths, Biology, Chemistry, Physics |
+
+Vector-L1-4B identifies itself as **Vector**, a teaching assistant by MikaLabs.
+
+---
+
+## Intended Use
+
+Vector-L1-4B is intended as a **teaching-assistant model** for educators. It is good at:
+
+- **Differentiated worksheets** — multi-tier (support / core / extension) question sets that show genuine difficulty progression.
+- **Mark schemes** — with method marks (M) and answer marks (A) shown separately.
+- **Misconception guides** — listing common, subject-specific student misconceptions and how to address them.
+- **Lesson plans** — structured with objectives, starters, main activities, and plenaries.
+- **Mixed-format questions** — short answer, true/false, fill-in-the-blank, calculation, explain-your-reasoning.
+- **Concept explanations** — pitched to a specified age or ability level.
+- **Following formatting and structural instructions** — e.g. "no multiple choice", "output as a markdown table", "give three tiers".
+
+### Out of Scope / Not Intended For
+
+- High-stakes or unsupervised assessment without a human teacher reviewing the output.
+- A substitute for a qualified teacher's judgement.
+- General-purpose chat, coding, or non-educational tasks (it is specialised).
+- Subjects outside Maths and the Sciences (coverage is weaker elsewhere).
+
+---
+
+## Strengths
+
+Vector-L1-4B punches well above its size as a teaching assistant. It excels at:
+
+- **Differentiated worksheets** with genuinely distinct support / core / extension tiers and real difficulty progression.
+- **Professional mark schemes** that separate method marks (M) from answer marks (A), the way real exam marking works.
+- **Subject-specific misconception guides** — identifying the actual errors students make on a topic and how to address them.
+- **Structured lesson plans** with clear objectives, starters, main activities, and plenaries.
+- **A wide range of question formats** — short answer, true/false with justification, fill-in-the-blank, calculation, and explain-your-reasoning — without defaulting to multiple choice.
+- **Strong instruction-following** on complex, multi-part requests (e.g. "three tiers, a mark scheme, misconceptions, no multiple choice, output as markdown").
+- **Accurate level calibration**, pitching difficulty appropriately for the age or ability you specify.
+- **Clean, ready-to-use output** — it produces the resource you asked for directly, without conversational filler.
+
+## A Note on Scale
+
+Vector-L1-4B is a compact 4-billion-parameter model designed to run on everyday school hardware. It is built for **school and secondary-level teaching**, not university or research-level material. On very hard problems it may occasionally make mistakes, so — as with any AI tool — **answer keys and factual content should be reviewed by a teacher before use with students.**
+
+## How to Use
+
+Example (transformers):
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_id = "MikaLabs/Vector-L1-4B"
+tok = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
+
+messages = [
+    {"role": "user", "content": "Create a differentiated worksheet on Pythagoras' theorem for a mixed-ability class. Three tiers with 3 questions each, a mark scheme with method and answer marks, and a list of common misconceptions. No multiple choice."}
+]
+inputs = tok.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
+out = model.generate(inputs, max_new_tokens=2048, temperature=0.7)
+print(tok.decode(out[0][inputs.shape[-1]:], skip_special_tokens=True))
+```
+
+**Recommended generation settings:** temperature 0.7, top_p 0.8.
+
+---
+
+## Ethical Considerations & Responsible Use
+
+- Outputs — especially answer keys and scientific facts — **must be reviewed by a qualified educator** before use with students.
+- It is an assistant, not an authority.
+- It is specialised for English-language Maths and Science teaching; quality and accuracy degrade outside that scope.
+
+---
+
+## Citation
+
+```
+@misc{vector-l1-4b,
+  title  = {Vector-L1-4B: An Open Teaching-Assistant Model},
+  author = {MikaLabs},
+  year   = {2026},
+  url    = {https://huggingface.co/MikaLabs/Vector-L1-4B}
+}
+```
+
+## Acknowledgements
+
+Built on Qwen3-4B-Instruct-2507 by the Qwen team, used under the Apache 2.0 license.