NC Bench
Overview
About
Tests
Models
Model
Claude 3.0 Sonnet
Test
Dialogue tags
Scenario
Write unattributed dialogue
Run ID
2024-07-15-01J2TTHW7WAHR2NM2D51HGNRD1
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
"Morning, Elsie. The usual coffee and croissant?" "You know me too well, Martha. Though I might get a couple of those lemon tarts to go. Ned has a bit of a sweet tooth." "Say no more. I'll box them up for you." "Have you heard the latest gossip? Old Mrs. Withers claims she saw a suspicious figure lurking around the abandoned mill last night." "Again? That woman has an overactive imagination. Probably just a raccoon or something." "I don't know. With all the strange occurrences lately, I can't help but wonder." "Like what? That broken window at the library?" "And the missing garden gnomes from the Johnsons' yard. Something peculiar is afoot in our little town." "You're starting to sound like one of those amateur sleuths from your mystery novels." "Maybe I should look into it. Add a bit of excitement to my retirement years." "Be careful what you wish for, Elsie. The last thing we need is another mystery on our hands." "Oh, hush. It might be fun. Imagine the stories I could tell the grandkids."
id
"gen-i5XyrLn4oSKSBwFkKLrXaxxyJBmH"
total_cost
0.004251
upstream_id
"msg_01CZ5fQBZgPyVkmo7Ld58fVi"
created_at
"2024-07-15T09:11:27.68477+00:00"
Monday, July 15th, 2024 at 9:11:27 AM GMT+00:00
model
"anthropic/claude-3-sonnet:beta"
app_id
182717
streamed
false
cancelled
false
provider_name
"Anthropic"
latency
500
moderation_latency
(null)
generation_time
340
finish_reason
"end_turn"
tokens_prompt
69
tokens_completion
236
native_tokens_prompt
77
native_tokens_completion
268
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.004251
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)