Update README.md

README.md

@@ -15,10 +15,10 @@ tags:
 - hybrid
 ---
 
-# Pebble-10M-Base
+# Pebble-10M
 ![Banner](banner.png)
 
-Pebble-10M-Base is a compact, hybrid autoregressive language model. It combines the efficiency of state-space models with the proven performance of attention layers, optimized using a custom Muon + AdamW optimizer split.
+Pebble-10M is a compact, hybrid autoregressive language model. It combines the efficiency of state-space models with the proven performance of attention layers, optimized using a custom Muon + AdamW optimizer split.
  
 ## Model Details
  
@@ -48,7 +48,7 @@ The model was trained on a 25B token subset of the following datasets:
 
 ## Benchmarks
  
-Pebble-10M-Base performs above random chance on several commonsense and arithmetic benchmarks.
+Pebble-10M performs above random chance on several commonsense and arithmetic benchmarks.
  
 | Benchmark | Accuracy | Random Baseline |
 |-----------|----------|------------------|
@@ -84,7 +84,7 @@ Here is a simple Python script to load the model and generate text interactively
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
  
-MODEL_ID = "basically-ai/Pebble-10M-Base"
+MODEL_ID = "basically-ai/Pebble-10M"
  
 def main():
     print("Loading Pebble 10M...")