General
Tulu Human Evaluation
Released Tulu 1 human acceptability judgments for four language models on 332 instruction-following prompts.
1,028items
4subjects
Apache-2.0 for the original Open Instruct release; source prompts and model outputs retain their upstream terms.license
instruction_followingdomain
textmodality
item-level responses released
Saturation status: Yes
Response matrix
Loading response matrix…
Correct (1)Incorrect (0)Unobserved
Scale: 1 = correct · 0 = incorrect