INDEX

Explanations

words related to geographical locations or countries specifically focusing on neighboring relationships

oai_token-act-pair · gpt-3.5-turbo

instances of the word "neighboring" or "neighbouring"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

odor

-0.93

inen

-0.90

ocker

-0.82

istry

-0.79

endi

-0.78

anwhile

-0.76

icer

-0.75

erer

-0.75

apego

-0.74

unker

-0.74

POSITIVE LOGITS

 neighbors

1.00

 neighbor

0.97

 Territories

0.92

 neighbours

0.91

 neighbour

0.90

 territories

0.89

 neighboring

0.88

 neighbouring

0.85

 regions

0.81

 allied

0.81

Activations Density 0.012%

words related to geographical locations or countries specifically focusing on neighboring relationships

instances of the word "neighboring" or "neighbouring"

No Comments

No Known Activations