NC Bench
Overview
About
Tests
Models
Model
Llama 3.1 8B
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-23-01J3G81TEN9411W701DR7FMK5K
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 leaned in, a sly grin spreading across her face as she leaned against the dimly lit alleyway wall. "Okay, Bill, here's the plan. We wait until the birthday boy's parents are distracted by the piƱata, then we slip in and start grabbing presents." Bill rubbed his chin thoughtfully, his eyes glinting with excitement. "I like the part where we grab the presents, but what's the holdup on the distraction? Can't we just, I don't know, shoot a Nerf gun at the parent's ankles or something?" Molly rolled her eyes. "Bill, we're not 12-year-old hackers, we're sophisticated thieves. We need a real diversion, like a strategically-timed clown horn solo." Bill's face lit up. "Now you're talking. But what about the clown? I thought we were sticking to the original plan." "Which involved what, exactly?" Molly asked, flipping open her phone to scan the layout of the party. "Oh, right. No clown. Um, what's the plan again?" "The plan is you create a scene and I swipe the gifts. Got it?" Bill scratched his head. "That's it. I'm so confused."
id
"gen-olQ9RQsYBMJVnQIuO2vnJcB0qoVi"
total_cost
0.0000648
upstream_id
"00bf3348-d2d9-4cfc-8353-e6e0c3e3cd6e"
created_at
"2024-07-23T16:51:04.698411+00:00"
Tuesday, July 23rd, 2024 at 4:51:04 PM GMT+00:00
model
"meta-llama/llama-3.1-8b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
973
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
102
tokens_completion
253
native_tokens_prompt
70
native_tokens_completion
254
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0000648
Evaluation details
Result
Evaluator
Details
Meta Data
35.0022%
Matches word count
n/a
neededClean
false
words
182
0.0000%
Dialogue to Total Word Ratio
Ratio: 70.33%, Deviation: 19.67%
neededClean
false
wordsTotal
182
wordsDialogue
128
17.5011%