INDEX

Explanations

references to detectives, investigations, crime, and mysteries in a variety of contexts

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jri

-0.86

eworld

-0.81

xual

-0.80

enei

-0.71

idays

-0.70

ais

-0.70

edom

-0.69

tu

-0.65

si

-0.65

enes

-0.64

POSITIVE LOGITS

 Grimoire

0.96

ective

0.83

 detective

0.76

 Holmes

0.76

 Conan

0.72

 Detective

0.70

 Inspector

0.69

 CLSID

0.65

 Guild

0.65

 Goodwin

0.63

Activations Density 5.158%

references to detectives, investigations, crime, and mysteries in a variety of contexts

No Comments

No Known Activations