GSM8K
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GSM8K 5-shot
78.32Accuracy
12
GSM8K
2.3Latency (s)
12
GSM8K traces (held-out)
0FPR
12
GSM8K
68.2Average Score @10
12
GSM8K
160Training Steps to Target Accuracy
12
GSM8K (test)
91.5Accuracy (Avg@8)
12
GSM8K
100True Workspace Rate
12
GSM8K (test)
53.68S_ref
12
GSM8K
9,068Max Length
12
GSM8K (test)
81.88Absolute Score
12
GSM8K
90.4Accuracy (GSM8K)
12
GSM8K ID (test)
54.59Accuracy
12
GSM8K (test)
2.489Test Perplexity
12
GSM8K (train)
0.335Training Loss
12
GSM8K
41.17Accuracy (maj@4)
12
GSM8K
67.3Accuracy
12
GSM8K
96.6Accuracy
12
GSM8K
88.32Accuracy (GSM8K)
12
GSM8K
87.92Accuracy (GSM8K)
12
GSM8K
75.04Mean Value (MV)
12
GSM8K
96.31Top-1 Accuracy
12
GSM8K
93.21Accuracy
12
GSM8K 4-shot max_gen_len=256
79.15Accuracy (%)
12
GSM8K
86.8Accuracy
12
GSM8K
79.55Accuracy
12