Update README.md

README.md

@@ -3,9 +3,9 @@ base_model:
 - EleutherAI/pythia-160m
 ---
 # Model Card
-The Pythia-160M model is designed for research on language model behavior and interpretability, trained on the Pile dataset. It has been evaluated on tasks like hellaswag and can be fine-tuned for further experimentation.
+The Pythia-160M model is designed for research on language model behavior and interpretability, trained on the [Pile](https://pile.eleuther.ai/) dataset. It has been evaluated on tasks like hellaswag and can be fine-tuned for further experimentation.
 
-## Evaluation Results - Hellaswag
+## Hellaswag Eval
 Evaluated on the Eleuther evaluation harnessre, vision 100,000 steps
 
 |  Tasks  |Version|Filter|n-shot| Metric |   |Value |   |Stderr|
@@ -14,5 +14,4 @@ Evaluated on the Eleuther evaluation harnessre, vision 100,000 steps
 |         |       |none  |     0|acc_norm|↑  |0.3082|±  |0.0046|
 
 ## How to Use
-
-To use this model, simply download the checkpoint and load it into your preferred deep learning framework.
+Not intended for deployment or human-facing interactions.