Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Language Modeling on C4 LLaMA-130M (val)

18.504Perplexity

ALIAS Adam version

17.8873622.0496826.21230.37432Jun 4, 2025Jul 5, 2025Aug 5, 2025Sep 5, 2025Oct 6, 2025Nov 6, 2025Dec 8, 2025
Updated 2mo ago

Evaluation Results

MethodLinks
2025.06
18.504-2.918-
2025.06
18.634-2.925-
2025.06
18.672-2.927-
2025.06
18.727-2.93-
2025.06
18.897-2.939-
2025.06
19.693-2.98-
2025.06
19.923-2.992-
2025.06
20.169-3.006-
2025.06
20.169-3.006-
2025.06
20.422-3.017-
2025.06
20.537-3.022-
2025.06
20.791-3.035-
2025.06
20.923-3.041-
2025.06
20.923-3.041-
2025.12
22.510.54--
2025.12
22.580.5--
2025.12
22.750.63--
2025.12
22.860.8--
2025.09
22.940.79-134
2025.06
22.982-3.135-
2025.12
23.10.46--
2025.12
23.350.57--
2025.12
23.730.53--
2025.12
23.740.52--
2025.09
23.740.79-106
2025.12
23.830.46--
2025.12
23.840.46--
2025.09
24.310.67-90
2025.09
24.361-134
2025.09
24.480.7-94
2025.09
24.590.7-94
2025.09
25.290.86-134
2025.09
25.340.79-134
2025.09
25.360.79-134
2025.09
25.880.86-134
2025.09
26.040.72-97
2025.12
26.470.57--
2025.09
29.370.86-134
2025.12
30.020.52--
2025.09
33.920.86-134