INDEX

Explanations

api response resource

This neuron fires on snippets of programming code (keywords, punctuation, and structure typical of source code).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 bombe

-0.93

 отвер

-0.85

CRIP

-0.80

exitRule

-0.79

 slogan

-0.79

 hemi

-0.79

菌

-0.77

⚛

-0.77

 dien

-0.76

енных

-0.76

POSITIVE LOGITS

 response

1.01

api

0.85

 resource

0.85

API

0.85

api

0.85

 Response

0.84

resposta

0.80

 responses

0.79

response

0.79

 http

0.79

Activations Density 0.053%