INDEX

Explanations

Catholic-related words and concepts

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

cki

-0.63

Keefe

-0.58

fell

-0.57

enhagen

-0.54

bite

-0.53

culosis

-0.52

lihood

-0.52

ãģı

-0.51

 Tactics

-0.51

POSITIVE LOGITS

arine

0.88

alog

0.77

aline

0.74

alyst

0.73

olics

0.72

cart

0.71

alogue

0.70

ones

0.70

 Cath

0.70

ulhu

0.67

Activations Density 9.608%

Catholic-related words and concepts

No Comments

No Known Activations