NC Bench
Overview
About
Tests
Models
Model
Llama 3.1 405B
Test
Dialogue tags
Scenario
Write unattributed dialogue
Run ID
2024-07-23-01J3G4ZSN4ESPGF28R6WB3VBSK
Model input
User/Human
You are an expert novel writer in the cozy mystery genre. Write a scene of about 200 words, with dialogue between two characters. The dialogue should be unattributed, meaning the speaker is not explicitly mentioned. The scene should be set in a small town bakery. Only return the story, nothing else.
Model response
Messages
Meta
AI
The sweet aroma of freshly baked bread wafted through the air, enticing the senses of everyone who walked into the Small Town Sweets bakery. The bell above the door jingled, signaling the arrival of a customer. "I need a cake for the Historical Society's meeting tonight. Something that'll impress the new members." "What's the theme of the meeting?" "They're discussing the history of our town's railroad. Something with trains would be nice." "Hmm, that's a tough one. I think I have just the thing." A few moments of rummaging through a stack of cake decorating books yielded a triumphant cry. "How about a cake shaped like the old train depot?" "That sounds perfect! Can you have it ready by five?" "I can do that, but I have to warn you, it won't be cheap. I'll have to use special molds and edible images to get the details just right." "Money's no object. The Historical Society is willing to pay top dollar for something that'll make a real impression. Just make sure it tastes as good as it looks."
id
"gen-OCCqErejIqzMc4sB0u7CO4p4LljT"
total_cost
0.000888
upstream_id
"858fdac7-e210-45a1-83aa-91cb57a2b026"
created_at
"2024-07-23T15:57:39.075692+00:00"
Tuesday, July 23rd, 2024 at 3:57:39 PM GMT+00:00
model
"meta-llama/llama-3.1-405b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
7620
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
103
tokens_completion
223
native_tokens_prompt
71
native_tokens_completion
225
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.000888
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)