NC Bench
Overview
About
Tests
Models
Model
Gemma 2 9B
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35SNABFBB8ZWHX7N72YMN3Z
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
"Alright, Bill, listen up. This isn't some stuffy museum heist, we're talking presents. Stuffed animals, board games, the works. Basically, a sugar-fueled paradise." "You callin' me a sugar fiend, Molly?" "Don't get your knickers in a twist. Just sayin', we're gonna be swimming in loot. Now, the intel says there's a bouncy castle, right? We swing in, grab the goods, swing out." "Bouncy castle? You're serious? We're gonna look like a couple of clowns trying to escape on a pogo stick." "Suit yourself, Bill. I'm picturing a graceful pirouette, a subtle snatch, and bam! Gone before they even realize we're missing." "Graceful? You? You tripped over a dust bunny last week." "Hey, that was a strategically placed dust bunny, designed to throw off pursuers. Look, just trust me. We'll be in and out before the pinata explodes." "Fine, fine. But if I see a unicorn, I'm taking it. No questions asked." "Deal. Now, how about we scope out the perimeter and figure out our escape route? I'm thinking a strategically placed cupcake explosion."
id
"gen-VAERWsIPFm0teuiUabaiDsxxCzt6"
total_cost
0.0000706
upstream_id
"b766bcb1-862a-4f1b-b84d-93eea5e046d2"
created_at
"2024-07-19T15:27:13.449018+00:00"
Friday, July 19th, 2024 at 3:27:13 PM GMT+00:00
model
"google/gemma-2-9b-it"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
3926
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
80
tokens_completion
262
native_tokens_prompt
75
native_tokens_completion
278
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0000706
Evaluation details
Result
Evaluator
Details
Meta Data
1.0360%
Matches word count
n/a
neededClean
false
words
174
36.7879%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
174
wordsDialogue
174
18.9120%