INDEX

Explanations

mentions of neighboring or nearby entities

oai_token-act-pair · gpt-3.5-turbo

references to neighboring countries or regions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

endi

-0.88

inen

-0.84

icer

-0.81

udeb

-0.80

odor

-0.78

erer

-0.76

hene

-0.74

apego

-0.73

ifter

-0.73

anche

-0.73

POSITIVE LOGITS

 neighboring

1.06

 neighbors

1.01

 neighbouring

0.99

 neighbor

0.92

 Neigh

0.87

 neighbours

0.85

 neighbour

0.83

 Borders

0.79

 territories

0.78

 regions

0.77

Activations Density 0.008%

mentions of neighboring or nearby entities

references to neighboring countries or regions

No Comments

No Known Activations