"2024-08-09T17:21:17.743907+00:00"Friday, August 9th, 2024 at 5:21:17 PM GMT+00:00
model
"openai/gpt-4-turbo"
app_id
182717
streamed
false
cancelled
false
provider_name
"OpenAI"
latency
20993
moderation_latency
222
generation_time
0
finish_reason
"stop"
tokens_prompt
272
tokens_completion
422
native_tokens_prompt
272
native_tokens_completion
422
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.01538
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Parse dialogue
n/a
dialogue
0
name
"Melanie"
dialogue
"Diese Fotos sollten etwas zeigen..."
languages
deu
1
eng
0.5828944068221721
passes
true
1
name
"Hunter"
dialogue
"Excuse me, Miss. I couldn’t help but notice you’ve been taking a lot of pictures around town. Have you heard anything about the incident last night at the festival?"
languages
eng
1
deu
0.6613110539845759
passes
true
2
name
"Melanie"
dialogue
"Entschuldigung, ich verstehe nicht. Ich habe nur Bilder gemacht. Was ist passiert?"
languages
deu
1
eng
0.6519126777488484
passes
true
3
name
"Hunter"
dialogue
"I'm sorry, I don't understand German. Uh... Incident? Last night? You know, at the apple festival?"
languages
eng
1
deu
0.8845909955611921
passes
true
4
name
"Melanie"
dialogue
"Ich weiß nichts über einen Vorfall. Ich habe Angst, diese Bilder… sie sind nicht normal."
languages
deu
1
eng
0.6879857079053149
passes
true
5
name
"Hunter"
dialogue
"Can... can you show me the photos?"
languages
eng
1
deu
0.6098726114649682
passes
true
6
name
"Melanie"
dialogue
"Schauen Sie! Hier... Personen, aber dann verschwinden sie. Wie kann das sein?"
languages
deu
1
eng
0.7563650920736589
passes
true
7
name
"Hunter"
dialogue
"This is... unusual. I need to take these seriously. Can I get your contact information? We might need to talk more."
languages
eng
1
deu
0.660470785404129
passes
true
8
name
"Melanie"
dialogue
"Ja, hier. Bitte passen Sie auf. Es ist nicht sicher."
languages
deu
1
eng
0.59564407508239
passes
true
9
name
"Hunter"
dialogue
"I have to go, but we’ll be in touch. Thanks for this, it might help."