Update README.md

README.md

@@ -124,9 +124,11 @@ At 494.9M parameters trained from scratch on a single Apple Silicon machine, TCF
 
 | File | Description |
 |------|-------------|
-| `leeknet_500m.py` | Architecture |
-| `train_stage_a.py` | Stage A pretraining |
-| `train_stage_b.py` | Stage B instruction fine-tuning |
+  - stage_b_step15250_best.npz — 1.98 GB weights
+  - leeknet_500m.py — architecture
+  - chat_stage_b.py — chat interface
+  - tokenizer/leek_bpe_32k.model — SentencePiece model
+  - tokenizer/leek_bpe_32k.vocab — vocabulary
 
 ---