Update README.md

README.md

@@ -1,18 +1,19 @@
-# Model Card for t5-small-wikitext
-## Model Description
+# t5-small-wikitext
 
-# Training Details
+t5-small trained on [wikitext/wikitest-103-raw-v1](wikitext/wikitest-103-raw-v1) over 50k steps (around 2 hours of training) following [T5 paper](https://arxiv.org/pdf/1910.10683.pdf) training procedure.
 
-## Training Data
-## Training Procedure
-### Preprocessing
-### Speeds, Sizes, Times
+* batch_size: 32
+* max_seq_length: 128
+* optim: Adafactor
+* sheduler: inverse square root (10k warm-up steps)
 
 ---
-license: apache-2.0
-datasets:
-- wikitext
-language:
-- en
-pipeline_tag: text2text-generation
+- license:
+  apache-2.0
+- datasets:
+  wikitext
+- language:
+  en
+- pipeline_tag:
+  text2text-generation
 ---