INDEX

Explanations

words related to attractions or points of interest

oai_token-act-pair · gpt-3.5-turbo

terms related to tourist attractions

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

igans

-0.73

ordan

-0.72

Ð¼

-0.71

ennett

-0.70

nia

-0.70

Ñı

-0.66

monds

-0.66

 citizenship

-0.65

uckles

-0.65

POSITIVE LOGITS

 attraction

1.34

 attractions

1.21

ĸļ

0.96

 Flavoring

0.92

pole

0.75

rers

0.74

inson

0.70

 Spit

0.67

ãħĭ

0.67

bucks

0.67

Activations Density 0.022%

words related to attractions or points of interest

terms related to tourist attractions

No Comments

No Known Activations