Update README.md

README.md

@@ -1,59 +1,87 @@
 ---
-base_model: google/gemma-3-270m-it
+license: mpl-2.0
 library_name: transformers
-model_name: Spark-270M
 tags:
-- generated_from_trainer
-- trl
-- hf_jobs
-- sft
-licence: license
+- gemma-3
+- synthetic-data
+- textbooks
+- distillation
+- utility
+- summarization
+- lightning
+- conversational
+base_model: google/gemma-3-270m
+datasets:
+- TitleOS/Spark-Lightning-Synthetic-Textbooks
+language:
+- en
+pipeline_tag: text-generation
 ---
 
-# Model Card for Spark-270M
+# Spark-270M
 
-This model is a fine-tuned version of [google/gemma-3-270m-it](https://huggingface.co/google/gemma-3-270m-it).
-It has been trained using [TRL](https://github.com/huggingface/trl).
+**Spark-270M** is a highly compact, utility-focused language model with **270 million parameters**. It is a fine-tune of Google's [Gemma 3 270M](https://huggingface.co/google/gemma-3-270m), designed to punch significantly above its weight class by leveraging high-quality synthetic data distillation.
 
-## Quick start
+The model functions as a "dense information engine"—specializing in generating concise title summaries, search engine queries, and logical follow-up questioning—while retaining the creative conversational flair inherited from its teacher model's lineage.
 
-```python
-from transformers import pipeline
+## ⚡ Model Details
+
+- **Model Name:** Spark-270M
+- **Base Architecture:** [Google Gemma 3 270M](https://huggingface.co/google/gemma-3-270m)
+- **Parameters:** 270M active parameters
+- **Context Window:** 32k tokens
+- **Teacher Model:** Lightning-1.7B (Custom model fine-tuned on Hermes 3)
+- **Training Type:** Synthetic "Textbook" Distillation (SFT)
+
+## 📚 Training Methodology: "Textbooks Are All You Need"
+
+Spark-270M was trained using a distinct data pipeline inspired by the *Textbooks Are All You Need* (Microsoft Phi) research paper. 
+
+Instead of training on raw web scrapes, Spark-270M was fine-tuned exclusively on a series of **synthetic textbooks** generated by a larger parent model, **Lightning-1.7B**. 
 
-question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="TitleOS/Spark-270M", device="cuda")
-output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
-print(output["generated_text"])
-```
+### The Teacher: Lightning-1.7B
+The data generator, Lightning-1.7B, was itself fine-tuned on the [Hermes 3 dataset](https://huggingface.co/nousresearch/hermes-3-llama-3.1-8b). This lineage allows Spark-270M to inherit specific behavioral traits from Hermes 3—namely creativity, steerability, and a refusal to be "boring"—despite being distilled into a rigid textbook format.
 
-## Training procedure
+The synthetic data focused on:
+1.  **High-density reasoning chains:** Explaining complex topics in compressed formats.
+2.  **Utility Tasks:** Converting conversational fluff into actionable queries.
+3.  **Socratic Dialogue:** Modeling inquisitive follow-up questioning.
 
- 
+## 🛠️ Intended Use & Capabilities
+
+Spark-270M is designed to be a lightweight **Utility Model**. It is ideal for edge devices, rapid prototyping, or functioning as a specific "node" in a larger agentic system (e.g., the summarizer node or the query-generator node).
+
+### Primary Capabilities
+* **Dense Title Summarization:** Converting long conversation threads into information-dense, short titles or abstracts.
+* **Search Query Generation:** Formulating precise, keyword-rich search queries based on vague user input.
+* **Proactive Questioning:** Generating relevant follow-up questions to clarify user intent or deepen a topic.
+
+## 💻 Example Usage
+
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
+model_id = "TitleOS/Spark-270M"
 
-This model was trained with SFT.
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
 
-### Framework versions
+# Example: Generating a search query from a user problem
+input_text = """
+User: I need to fix my sink, it's leaking from the bottom pipe where the U-shape thing is.
+Task: Generate 3 search engine queries for this problem.
+Response:
+"""
 
-- TRL: 0.26.1
-- Transformers: 4.57.3
-- Pytorch: 2.9.1
-- Datasets: 4.4.1
-- Tokenizers: 0.22.1
+input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")
 
-## Citations
+outputs = model.generate(**input_ids, max_new_tokens=128)
+print(tokenizer.d ecode(outputs[0]))
 
 
+Quants:
 
-Cite TRL as:
-    
-```bibtex
-@misc{vonwerra2022trl,
-	title        = {{TRL: Transformer Reinforcement Learning}},
-	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
-	year         = 2020,
-	journal      = {GitHub repository},
-	publisher    = {GitHub},
-	howpublished = {\url{https://github.com/huggingface/trl}}
-}
-```
+Q4_K_M: https://huggingface.co/TitleOS/Spark-270M-FP16-Q4_K_M-GGUF
+Q8: https://huggingface.co/TitleOS/Spark-270M-FP16-Q8_0-GGUF
+FP16: https://huggingface.co/TitleOS/Spark-270M-FP16
+Adaptor: https://huggingface.co/TitleOS/Spark-270M-LoRA