Update README.md

README.md

@@ -20,7 +20,7 @@ model-index:
           name: "Language Modeling"
           type: "text-generation"
         dataset:
-          name: "Tigrinya News Corpus (59 M tokens)"
+          name: "Tigrinya News Corpus (~60 M tokens)"
           type: "text"
           split: "validation"
         metrics:
@@ -40,7 +40,7 @@ This model demonstrates good generation and completion capabilities for Tigrinya
 
 ## Model Details
 
-- **Model Type:** Causal Language Model (Autoregressive)
+- **Model Type:** Causal Language Model
 - **Base Model:** [google/gemma-3-4b-pt](https://huggingface.co/google/gemma-3-4b-pt)
 - **Parameters:** 4 billion
 - **Architecture:** Gemma 3 with `Gemma3ForCausalLM`
@@ -157,7 +157,7 @@ The following examples demonstrate the model's capabilities across different con
 | Training loss   | train      | 0.48     |
 | Validation loss | validation | 0.91     |
 
-*Validation corpus* – held-out 3 % of a 59 M-tokens.
+*Validation corpus* – held-out 3 % of a ~60 M-tokens.
 
 ## Limitations