# Foundation-Sec-8B

Foundation-Sec-8B (Llama-3.1-FoundationAI-SecurityLLM-base-8B) is an open-weight, 8-billion parameter base language model specialized for cybersecurity applications. It extends Llama-3.1-8B model through continued pretraining on a curated corpus of cybersecurity-specific text, including threat…

- **Kind:** model (Hugging Face)
- **Hugging Face:** https://huggingface.co/fdtn-ai/Foundation-Sec-8B
- **Owner:** fdtn-ai (https://forefy.com/u/hf/fdtn-ai)
- **Parameters:** 8.0B
- **Context:** 131072 tokens
- **License:** apache-2.0
- **Focus:** defensive
- **Architecture:** LlamaForCausalLM
- **Pinned revision:** `9e6bb1f70b402e55bbf68cfd7abf88458daf09b3`
- **Page:** https://forefy.com/models/b31b8377-4111-55aa-8eeb-618024c8cd76
- **API (JSON):** https://forefy.com/api/asr/models/b31b8377-4111-55aa-8eeb-618024c8cd76

## What was done to it

- finetune (declared evidence, from meta-llama/Llama-3.1-8B)

## Lineage

meta-llama/Llama-3.1-8B → fdtn-ai/Foundation-Sec-8B

## Checks (repo at the pinned revision; not a behavior test)

- **Identity** [pass]: No lookalike name or conflicting provenance
- **Weights format** [pass]: safetensors only, nothing runs on load
- **Pickle files** [pass]: None.
- **Remote code** [pass]: None
- **Chat template** [info]: None shipped
- **Refusals modified** [pass]: No
- **License** [warn]: Declares apache-2.0, but derives from a llama3.1 model whose terms carry over
- **Lineage** [pass]: Derives from meta-llama/Llama-3.1-8B
- **Model card** [pass]: 5 of 6, missing evaluation results
- **Safetensors headers** [pass]: Every byte is indexed, no hidden data

## Variants

- BF16 safetensors (safetensors, 16.1 GB)

## Run it

- transformers: `AutoModelForCausalLM.from_pretrained("fdtn-ai/Foundation-Sec-8B", revision="9e6bb1f70b402e55bbf68cfd7abf88458daf09b3")`