update readme

README.md

@@ -20,7 +20,7 @@ Trained on a corpus of 4 trillion tokens, this model demonstrates that native 1-
 
 ➡️ **Technical Report:** [BitNet b1.58 2B4T Technical Report](https://arxiv.org)
 
-➡️ **Official Code:** [microsoft/BitNet (bitnet.cpp)](https://github.com/microsoft/BitNet)
+➡️ **Official Inference Code:** [microsoft/BitNet (bitnet.cpp)](https://github.com/microsoft/BitNet)
 
 ## Model Variants
 
@@ -112,7 +112,6 @@ BitNet b1.58 2B4T was evaluated against leading open-weight full-precision LLMs
 | **Latency (CPU Decoding)** | 48ms         | 41ms       | 65ms         | 67ms         | 124ms      | **29ms** |
 | **Energy (Estimated)** | 0.258J       | 0.186J     | 0.347J       | 0.425J       | 0.649J     | **0.028J** |
 | **Training Tokens (Pre-train)**| 9T* | 2T** | 18T          | 11T          | 1.1T       | 4T                  |
-|--------------------------------|--------------|------------|--------------|--------------|------------|---------------------|
 | ARC-Challenge   | 37.80        | 38.40      | 46.67        | 43.52        | 44.80      | **49.91** |
 | ARC-Easy        | 63.17        | 63.13      | **76.01** | 62.92        | 72.14      | 74.79               |
 | OpenbookQA      | 34.80        | 38.80      | 40.80        | **46.00** | 40.20      | 41.60               |