Update README.md

README.md

@@ -56,7 +56,7 @@ Puffin and Hermes-2 both beat previous SOTA for GPT4ALL benchmarks, with Hermes-
 
 - Hermes 2 is trained on purely single turn instruction examples.
 
-- Puffin is trained mostly on long multi-turn, long context GPT-4 conversations, as well as curated single-turn examples relating to Physics, Bio, Math and Chem.
+- Puffin is trained mostly on multi-turn, long context, highly curated and cleaned GPT-4 conversations with real humans, as well as curated single-turn examples relating to Physics, Bio, Math and Chem.
 
 For these reasons, it's reccomended to give Puffin a try if you want to have multi-turn conversations and/or long context communication.
 
@@ -173,5 +173,4 @@ AGI Eval:
 |agieval_sat_en_without_passage|      0|acc     |0.4223|±  |0.0345|
 |                              |       |acc_norm|0.3447|±  |0.0332|
 |agieval_sat_math              |      0|acc     |0.3364|±  |0.0319|
-|                              |       |acc_norm|0.2773|±  |0.0302|```
-
+|                              |       |acc_norm|0.2773|±  |0.0302|```