Datasets:

wonabru-org
/

simple_test_predictable_llm_tokens

Modalities:

Formats:

Size:

Libraries:

Dataset card Data Studio Files Files and versions Community

wonabru commited on Feb 3

Commit

a60202a

·

verified ·

1 Parent(s): 0a7b77e

Update README.md

Files changed (1) hide show

README.md +1 -0

README.md CHANGED Viewed

@@ -15,3 +15,4 @@ configs:
   - split: train
     path: data/train-*
 ---

   - split: train
     path: data/train-*
 ---
+Artificial data which comes from simple sinus function of 16k tokens vocab size. Only for sanity testing if model predicts anything. Amplitudes and frequencies were randomized.