INDEX

Explanations

references to specific places or locations

oai_token-act-pair · gpt-3.5-turbo

words related to specific names or entities, particularly proper nouns

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

WAR

-0.72

gage

-0.72

kies

-0.70

dies

-0.70

venge

-0.67

CAP

-0.66

gard

-0.66

MET

-0.65

sea

-0.65

 è£ı

-0.65

POSITIVE LOGITS

imb

1.13

odies

0.91

ilib

0.84

oche

0.81

ead

0.79

orts

0.78

uilt

0.78

ospital

0.77

edience

0.76

amate

0.75

Activations Density 0.012%

references to specific places or locations

words related to specific names or entities, particularly proper nouns

No Comments

No Known Activations