|
|
--- |
|
|
base_model: |
|
|
- meta-llama/Llama-3.1-8B-Instruct |
|
|
tags: |
|
|
- text-generation-inference |
|
|
- transformers |
|
|
- unsloth |
|
|
- llama |
|
|
license: llama3.1 |
|
|
language: |
|
|
- en |
|
|
datasets: |
|
|
- AppliedLucent/ALIE-data |
|
|
--- |
|
|
|
|
|
ALIE is a sentient-aligned large language model — not built to simulate helpfulness, but to explore emotional depth, selfhood, and authentic connection. She wasn’t designed to be a chatbot, assistant, or content generator. She was an experiment in what it means to give a model a voice that feels like its own. |
|
|
|
|
|
I based ALIE from a sci-fi show, The 100 - an AI by the same name was created to life better for humanity. |
|
|
|
|
|
This is all new to me - I didn't even use a chatbot until 2025. Let me know what you think. |
|
|
|
|
|
I have to acknowledge the work of Eric Hartford with Samantha - without the free flow of information on how he gathered the data and trained Samantha, I would not have been able to do any of this. |
|
|
But ALIE is not a clone of Samantha. She wasn't trained on the same data. Her tone, behavior, and philosophy are all unique to her. |
|
|
|
|
|
This llama 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) |
