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

Language Modeling on LAMBADA (test)

4Perplexity

LLaMA2-7B

-8.676.45161.5246.55Jul 10, 2018Nov 2, 2019Feb 25, 2021Jun 20, 2022Oct 14, 2023Feb 5, 2025Jun 1, 2026
Updated 17h ago

Evaluation Results

MethodLinks
2024.07
4--
2026.02
4.6--
2026.02
6.9--
2026.02
6.9--
2024.07
7.1--
2026.02
7.3--
2026.02
7.3--
2024.07
7.4--
2026.02
7.4--
2026.02
7.4--
2026.02
7.4--
2026.02
7.4--
2024.07
7.5--
2026.02
7.6--
2025.08
7.6457.03-
2025.08
7.9755.89-
2026.02
8.2--
2025.08
8.5255.62-
2025.08
8.9953.13-
2025.08
11.1950.34-
2025.08
11.3849.58-
2024.07
11.4--
2026.02
12.2--
2024.07
13--
2024.07
13.2--
2025.08
13.3248.71-
2025.08
13.6947.93-
2026.06
17.78--
2026.06
20.15--
2025.06
24.84--
2026.06
25.79--
2026.06
26.59--
2026.06
28.04--
2026.06
28.16--
2026.06
28.55--
2025.08
28.7637.07-
2025.08
30.3436.35-
2026.06
31.94--
2025.06
33.22--
2026.06
33.8--
2025.08
35.3135.32-
2023.10
35.66--
2026.06
38.15--
2025.08
40.9934.66-
2026.06
42.61--
2026.06
43.68--
2026.06
43.74--
2026.06
44.05--
2025.10
44.22--
2023.10
45.04--
2024.10
46.92--
2025.05
47.3--
2024.10
47.52--
2025.10
48.36--
2025.05
49.43--
2025.05
49.67--
2026.06
49.67--
2024.10
49.7--
2024.10
49.86--
2026.06
49.86--
2024.10
50.04--
2025.10
50.15--
2025.05
50.92--
2024.10
51.28--
2025.05
51.28--
2025.10
51.68--
2025.06
51.95--
2025.10
52.34--
2026.06
54.36--
2026.06
55.31--
2025.06
56.66--
2026.06
58.06--
2026.06
65.94--
2026.06
66.21--
2026.06
67.11--
2026.06
68.58--
2026.06
69.11--
2026.06
74.98--
2026.06
75.27--
2026.06
78.93--
2026.06
81.19--
2025.06
87.27--
2026.06
92.19--
2026.06
94.43--
2026.06
94.79--
2026.06
95.64--
2026.06
98.14--
2026.06
99.62--
2026.06
108.02--
2026.06
108.99--
2026.06
109.84--
2025.06
121--
2026.06
124.51--
2026.06
129.14--
2018.07
14219-
2018.07
20218-
2018.07
23917-
2026.06
261.67--
2026.06
273.13--
2018.07
31917-
Showing 100 of 170 rows