INDEX

Explanations

names and words related to geographical locations, wars, military groups, and political activities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Examiner

-0.72

 attribution

-0.65

cientious

-0.64

swick

-0.64

 Excellence

-0.64

 Participant

-0.63

 Merchants

-0.63

vou

-0.61

 Zimmer

-0.61

 behavi

-0.59

POSITIVE LOGITS

ield

1.04

Å«

0.98

ielding

0.96

ank

0.94

cling

0.87

olk

0.86

enne

0.86

orkshire

0.86

uki

0.85

acht

0.85

Activations Density 7.234%

names and words related to geographical locations, wars, military groups, and political activities

No Comments

No Known Activations