Upload README.md with huggingface_hub

README.md

@@ -51,7 +51,7 @@ pipeline_tag: feature-extraction
 | **Vocabulary** | 100,000 whole words |
 | **Training data** | ~2B tokens (OpenWebText subset, DFSG-compliant) |
 | **Training hardware** | Single NVIDIA RTX 3090 |
-| **Training time** | ~24 hours (2M steps) |
+| **Training time** | ~4 days (2M steps) |
 | **License** | GPL-3.0 |
 | **Parameters** | 60M (30M target + 30M context embeddings) |
 
@@ -142,9 +142,9 @@ Full training code and visualizations: [github.com/ruapotato/Free-Language-Embed
 
 ## Interactive Visualizations
 
-- [Embedding Spectrogram](https://ruapotato.github.io/chat_hamner/spectrogram.html) — PCA waves, sine fits, cosine surfaces
-- [3D Semantic Directions](https://ruapotato.github.io/chat_hamner/semantic_3d.html) — See how semantic axes align in the learned geometry
-- [Training Dashboard](https://ruapotato.github.io/chat_hamner/dashboard.html) — Loss curves and training metrics
+- [Embedding Spectrogram](https://ruapotato.github.io/Free-Language-Embeddings/spectrogram.html) — PCA waves, sine fits, cosine surfaces
+- [3D Semantic Directions](https://ruapotato.github.io/Free-Language-Embeddings/semantic_3d.html) — See how semantic axes align in the learned geometry
+- [Training Dashboard](https://ruapotato.github.io/Free-Language-Embeddings/dashboard.html) — Loss curves and training metrics
 
 ## Citation