NC Bench
Overview
About
Tests
Models
Model
Llama 3 70B
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35T7HEZNMAJZK3JFC2RVAYQ
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
"I'm telling you, Bill, this is a goldmine. A room full of sugared-up kids and oblivious parents. We're in and out with the loot in under ten minutes." "Unless we get caught by the clown," Bill said, raising an eyebrow. "I'm not exactly eager to get a pie in the face or a squirt gun to the ear." "Oh, come on, Bill. We're professionals. We can totally take down a clown. Besides, it's not like they're packing heat. Worst case scenario, we get a hairy nose and a balloon animal." "And what's the plan for the presents, exactly? We can't exactly fit a bunch of wrapped boxes into our backpacks." "Leave that to me, my friend. I've got a system. We're going to swap out the gifts with fake ones we've been working on. The kids will be none the wiser, and we'll be rolling in Pokémon cards and Hatchimals." "Pokemon cards? You know the kids are going to notice when they open up a pack of 1998 reprints, right?" "Shh, don't ruin the magic, Bill. It's all about the thrill of the heist."
id
"gen-QzMLhxYySMjldnL69ImLX6225EXn"
total_cost
0.0002448
upstream_id
"cmpl-5c1c6d42e12349e1b35e7e734be3b830"
created_at
"2024-07-19T15:37:10.650753+00:00"
Friday, July 19th, 2024 at 3:37:10 PM GMT+00:00
model
"meta-llama/llama-3-70b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"OctoAI"
latency
4003
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
102
tokens_completion
248
native_tokens_prompt
72
native_tokens_completion
248
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0002448
Evaluation details
Result
Evaluator
Details
Meta Data
60.2752%
Matches word count
n/a
neededClean
false
words
185
75.3096%
Dialogue to Total Word Ratio
Ratio: 97.30%, Deviation: 7.30%
neededClean
false
wordsTotal
185
wordsDialogue
180
67.7924%