Update README.md

README.md

@@ -3,31 +3,13 @@ tags:
 - generated_from_trainer
 datasets:
 - openwebtext
-model-index:
-- name: llama-2-7b-hf-distill-kth-len256-random-shift-4-lr1e-5-decayto0
-  results: []
+license: llama2
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# llama-2-7b-hf-distill-kth-len256-random-shift-4-lr1e-5-decayto0
-
-This model is a fine-tuned version of [/scr-ssd/cygu/weights/Llama-2-7b-hf/](https://huggingface.co//scr-ssd/cygu/weights/Llama-2-7b-hf/) on the openwebtext dataset.
-
 ## Model description
 
-More information needed
-
-## Intended uses & limitations
-
-More information needed
+Logits-based watermark distilled Llama 2 7B using the KTH \\(s=4\\) watermarking strategy in the paper [On the Learnability of Watermarks for Language Models](https://arxiv.org/abs/2312.04469).
 
-## Training and evaluation data
-
-More information needed
-
-## Training procedure
 
 ### Training hyperparameters
 
@@ -45,13 +27,9 @@ The following hyperparameters were used during training:
 - lr_scheduler_warmup_steps: 500
 - training_steps: 5000
 
-### Training results
-
-
-
 ### Framework versions
 
 - Transformers 4.29.2
 - Pytorch 2.0.1+cu117
 - Datasets 2.13.1
-- Tokenizers 0.13.3
+- Tokenizers 0.13.3