Document mirrored Llama metadata and CogACT training base

README.md

@@ -50,9 +50,16 @@ checkpoints/frozen-ema-step-20000.pt
 
 Set `MEMORYVLA_SKIP_BASE_WEIGHTS=1` for inference. The task checkpoint already
 contains the trained LLM, vision backbone, projector, action model, and EMA
-weights. The loader still needs the Llama-2 tokenizer/config metadata from
-`NousResearch/Llama-2-7b-hf`; it does not need to download the full Llama or
-vision weight files again.
+weights. The required Llama-2 tokenizer/config metadata is mirrored at:
+
+```text
+base_models/llama-2-7b-hf/
+```
+
+It does not need to download the full Llama or vision weight files again.
+Point `MEMORYVLA_LLAMA2_7B_PATH` at that downloaded metadata directory for
+offline inference. The metadata is provided for loader compatibility; the
+gated Llama model license still applies to the underlying model family.
 
 For future real-task training, the matching CogACT-Large initialization
 checkpoint is available at:
@@ -61,6 +68,10 @@ checkpoint is available at:
 base_models/CogACT-Large/CogACT-Large.pt
 ```
 
+The CogACT base checkpoint is only needed when initializing a new training
+run. It is not needed when loading the Frozen+EMA task checkpoint for
+inference.
+
 The original Llama model weights are not duplicated in this repository because
 they are gated third-party weights. Use the original Hugging Face model with
 the appropriate access terms and token.