Closed impact-rm closed 3 months ago
When I call the eq-bench with -revise enabled it returns "! Error: 4 emotions were not returned" and revised_answers: {}
The problem is solved by increasing the COMPLETION_TOKENS (which is set to 60) in eq_bench_utils.py. As it's the value parsed to max_tokens of openai.
Ah, thank you. I overlooked this, will fix.
When I call the eq-bench with -revise enabled it returns "! Error: 4 emotions were not returned" and revised_answers: {}