NC Bench
Overview
About
Tests
Models
Model
AI21 Jamba
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35T8DRND5TJ994BZBKVZYZF
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% words of dialogue between two thieves named Molly and Bill who are planning to break into a children's party to steal all the presents. Only include the prose of the scene without any markdown or commentary.
Model response
Messages
Meta
AI
Molly: "Alright, let's go over the plan again. We'll sneak in during the birthday song, while all the parents are distracted." Bill: "Got it. But what about the kids? They're not exactly known for keeping quiet." Molly: "Exactly. That's why we'll have to be quick and quiet. No noise, no mess." Bill: "But how are we going to carry all those presents? They're usually pretty big and bulky." Molly: "We'll bring a couple of large bags and stuff them as full as we can. And if we can't fit everything, we'll make two trips." Bill: "But won't that be suspicious? Two grown-ups carrying bags full of presents out of a children's party?" Molly: "Not if we act like we're supposed to be there. We'll wear costumes, like we're entertainers or something." Bill: "Entertainers? What kind of entertainers? Clowns? Magicians? Balloon animals?" Molly: "Does it matter? As long as we look the part and get out of there before anyone suspects anything." Bill: "Fine, but I'm not wearing a tutu." Molly: "Deal. Let's do this."
id
"gen-4PLRavGXXJRKfulh2PQpmspX8M3G"
total_cost
0.0003167
upstream_id
"cmpl-25d483213e5042b4a865d3001f91770a"
created_at
"2024-07-19T15:37:40.716535+00:00"
Friday, July 19th, 2024 at 3:37:40 PM GMT+00:00
model
"ai21/jamba-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"AI21"
latency
1500
moderation_latency
(null)
generation_time
180
finish_reason
"stop"
tokens_prompt
68
tokens_completion
259
native_tokens_prompt
212
native_tokens_completion
301
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0003167
Evaluation details
Result
Evaluator
Details
Meta Data
2.0116%
Matches word count
n/a
neededClean
false
words
175
98.1147%
Dialogue to Total Word Ratio
Ratio: 93.71%, Deviation: 3.71%
neededClean
false
wordsTotal
175
wordsDialogue
164
50.0632%