Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -22,6 +22,8 @@ tags:
|
|
| 22 |
|
| 23 |
This is the first public release of an ongoing research project: the dataset was written from scratch, the LoRA was trained locally, and the architecture was modified and experimented with. The model is distributed as a **GGUF (Q4_K_M)** file ready to run in [Ollama](https://ollama.com).
|
| 24 |
|
|
|
|
|
|
|
| 25 |
## Why this model exists
|
| 26 |
|
| 27 |
Most fine-tuned models are optimized for one thing: following instructions well. 0pen is interested in something slightly different — being a **consistent presence in a conversation**. That means:
|
|
|
|
| 22 |
|
| 23 |
This is the first public release of an ongoing research project: the dataset was written from scratch, the LoRA was trained locally, and the architecture was modified and experimented with. The model is distributed as a **GGUF (Q4_K_M)** file ready to run in [Ollama](https://ollama.com).
|
| 24 |
|
| 25 |
+
**Source code & full release:** [github.com/0penAGI/0pen](https://github.com/0penAGI/0pen) — dataset pipeline, AGR training wrapper, adapter weights, and scripts.
|
| 26 |
+
|
| 27 |
## Why this model exists
|
| 28 |
|
| 29 |
Most fine-tuned models are optimized for one thing: following instructions well. 0pen is interested in something slightly different — being a **consistent presence in a conversation**. That means:
|