| license: agpl-3.0 | |
| library_name: pytorch | |
| tags: | |
| - deprecated | |
| - renamed | |
| pipeline_tag: text-generation | |
| # GlubLM 18M (DEPRECATED - renamed to 36M) | |
| > [!WARNING] | |
| > This repo has been renamed to [**DenSec02/glublm-36m**](https://huggingface.co/DenSec02/glublm-36m) because the model is actually 36M parameters. The `18m` slug was legacy from the initial v0.1.x experiment when the model was genuinely 18M params. | |
| All current weights, training data, model card, and documentation now live at: | |
| **https://huggingface.co/DenSec02/glublm-36m** | |
| Please update your code to the new slug: | |
| ```python | |
| # old: | |
| hf_hub_download("DenSec02/glublm-18m", ...) | |
| # new: | |
| hf_hub_download("DenSec02/glublm-36m", ...) | |
| ``` | |
| This repo is kept as a redirect marker. No files here are guaranteed to match the current checkpoint - always use `glublm-36m` for new work. | |
| - Source: https://github.com/Den-Sec/glublm | |
| - Dataset: https://huggingface.co/datasets/DenSec02/glublm-60k-ted | |
| - License: AGPL-3.0 | |