INDEX

Explanations

proper names or entities related to politics, crime, and individuals discussing or interacting with them

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

nces

-0.96

xual

-0.91

 Bengal

-0.91

bred

-0.89

fare

-0.85

Demand

-0.84

breeding

-0.82

 AMERICA

-0.81

 Liberia

-0.81

yrinth

-0.80

POSITIVE LOGITS

kson

2.49

sson

1.47

 Garner

1.17

herty

1.04

indal

1.03

sonian

0.99

Boe

0.99

istry

0.98

asure

0.97

 GOODMAN

0.97

Activations Density 1.271%

proper names or entities related to politics, crime, and individuals discussing or interacting with them

No Comments

No Known Activations