---
language: [en]
license: mit
tags:
- data-science
- machine-learning
- deep-learning
- statistics
- slm
- llama-style
- rope
- 5m-context
- from-scratch
- 1b-params
pipeline_tag: text-generation
---

# Data Scientist-SLM: Role-Based Small Language Model

A **LLaMA-style transformer** (~989.7M params, ~0.99B) trained from scratch for the **Data Scientist** role.
Supports up to **5M token context** via RoPE with gradient checkpointing.

## Architecture
| Component | Value |
|-----------|-------|
| Architecture | LLaMA-style (RoPE + RMSNorm + SwiGLU) |
| Parameters | ~989.7M (~0.99B) |
| Layers | 32 |
| Heads | 20 |
| Embedding | 1600 |
| Max Context | 5,000,000 tokens |
| Max Output | 5,000,000 tokens |
| Vocab | 2,025 BPE |
| Model Size | ~4 GB (fp32) |

## Training
- Best eval loss: 1.4600185036659241
- Trained with gradient checkpointing on Apple M4 (MPS)
- 3 epochs, batch_size=1, grad_accum=16

## Usage
```python
from huggingface_hub import hf_hub_download
from tokenizers import Tokenizer

model_path = hf_hub_download("sathishphdai/data-scientist-slm-5m", "model.safetensors")
tokenizer_path = hf_hub_download("sathishphdai/data-scientist-slm-5m", "data_scientist_tokenizer.json")
tokenizer = Tokenizer.from_file(tokenizer_path)
```