Cleaned, tokenized, and DataLoader-ready 1D linear algebra TRAINING dataset from DeepMind; for use with FLAN-T5

README.md

@@ -0,0 +1,21 @@
+---
+dataset_info:
+  features:
+  - name: input_ids
+    sequence: int32
+  - name: attention_mask
+    sequence: int8
+  - name: labels
+    sequence: int64
+  splits:
+  - name: train
+    num_bytes: 387376552
+    num_examples: 1999998
+  download_size: 59495383
+  dataset_size: 387376552
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+---

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e9629086e19c2723e531dd8aea97b8256a6df25c7baa8cc3de0dd64276c8612
+size 59495383