Update usage for DeepJ 0.13.0-alpha

Describe the current neural network library and retain model loading requirements, weights, and evaluation details.

README.md

@@ -18,11 +18,11 @@ DeepJ is a Java neural network library with shared tensors, automatic differenti
 
 This compact language model was created with [**DeepJ**](https://github.com/KirstenAli/DeepJ) and uses its Prism decoder-only Transformer architecture with 4 layers, a hidden size of 128, 4 attention heads, Q rank 64, KV rank 32, and a 2,048-token vocabulary.
 
-Generation uses a request-scoped key/value cache. By default, DeepJ rebuilds it when the context window shifts. Optional [rolling generation](https://github.com/KirstenAli/DeepJ/blob/v0.12.0-alpha/docs/inference.md#rolling-generation) avoids repeated rebuilds, with different attention behaviour after eviction; it does not extend the model's trained context capability. The `model.dj` checkpoint and `tokenizer.bpe` files use DeepJ's versioned binary formats. They are not directly loadable by the Python Transformers library. This older checkpoint predates embedded architecture settings, so its configuration must be supplied when loading it.
+Generation uses a request-scoped key/value cache. By default, DeepJ rebuilds it when the context window shifts. Optional [rolling generation](https://github.com/KirstenAli/DeepJ/blob/master/docs/inference.md#rolling-generation) avoids repeated rebuilds, with different attention behaviour after eviction; it does not extend the model's trained context capability. The `model.dj` checkpoint and `tokenizer.bpe` files use DeepJ's versioned binary formats. They are not directly loadable by the Python Transformers library. This older checkpoint predates embedded architecture settings, so its configuration must be supplied when loading it.
 
 ## Usage
 
-DeepJ `0.13.0-alpha` is prepared for release. Maven Central publication is pending; after publication, use:
+Use [DeepJ 0.13.0-alpha](https://github.com/KirstenAli/DeepJ):
 
 ```xml
 <dependency>
@@ -32,7 +32,7 @@ DeepJ `0.13.0-alpha` is prepared for release. Maven Central publication is pendi
 </dependency>
 ```
 
-Until publication, use the currently published [DeepJ 0.12.0-alpha](https://github.com/KirstenAli/DeepJ/releases/tag/v0.12.0-alpha) with this checkpoint, or build the prepared 0.13.0-alpha source from the DeepJ repository. This library release does not retrain or replace these model weights.
+DeepJ 0.13.0-alpha expands the library’s network APIs. These model weights and their loading requirements are unchanged.
 
 Download this repository's `model.dj` and `tokenizer.bpe` into `downloaded-model`. Imports are omitted below.