INDEX

Explanations

terms related to being blind or lacking awareness

oai_token-act-pair · gpt-3.5-turbo

references to blindness or ignorance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xit

-1.02

 Occupations

-0.74

OPLE

-0.74

AIN

-0.70

ICAN

-0.69

ALLY

-0.68

ickr

-0.67

HOU

-0.62

CE

-0.61

ATIVE

-0.61

POSITIVE LOGITS

fold

1.25

blind

1.22

sight

1.22

sided

1.20

 blind

1.04

ibility

1.01

ingly

0.93

ening

0.91

ison

0.87

sighted

0.87

Activations Density 0.014%

terms related to being blind or lacking awareness

references to blindness or ignorance

No Comments

No Known Activations