Add root config.json to enable Hub download tracking; document it in README

README.md

@@ -41,6 +41,7 @@ time-series encoder onto a pretrained LLM.
 | `stage-2/` | **Stage 2** | Stage 1 + a short alignment stage that interleaves the two modalities (best multimodal results); context extended to 4096. `model.safetensors` + `config.json` |
 | `tokenizer/` | — | 131k-vocabulary BPE tokenizer (trained from scratch) |
 | `model.py` · `tokenizer.py` | — | Minimal inference implementation (`Chronicle`, `ChronicleConfig`, `ChronicleTokenizer`) |
+| `config.json` | — | Repo-level model metadata (mirrors the stage-2 architecture); the per-stage `config.json` files remain authoritative for loading |
 
 ## Architecture
 

config.json

@@ -0,0 +1,15 @@
+{
+  "model_type": "chronicle",
+  "architectures": ["Chronicle"],
+  "checkpoints": ["stage-1", "stage-2"],
+  "default_checkpoint": "stage-2",
+  "sequence_len": 4096,
+  "vocab_size": 131072,
+  "n_layer": 16,
+  "n_head": 8,
+  "n_kv_head": 4,
+  "n_embd": 1024,
+  "patch_len": 32,
+  "num_quantiles": 21,
+  "torch_dtype": "bfloat16"
+}