INDEX

Explanations

references to admiration or worship, particularly of individuals

oai_token-act-pair · gpt-3.5-turbo

references to "idol" and its variants in various contexts

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

-0.79

RAW

-0.70

~~~~~~~~~~~~~~~~

-0.67

llular

-0.66

 Dull

-0.65

raq

-0.65

uckles

-0.65

arnaev

-0.65

esome

-0.64

TPS

-0.62

POSITIVE LOGITS

 idol

1.05

ãħĭ

0.93

 idols

0.84

 worshipped

0.84

imation

0.83

 worsh

0.83

 Idol

0.82

ãħĭãħĭ

0.82

imates

0.80

 worship

0.77

Activations Density 0.018%

references to admiration or worship, particularly of individuals

references to "idol" and its variants in various contexts

No Comments

No Known Activations