INDEX

Explanations

references to specific names, locations, and medical conditions

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names associated with historical figures and locations

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

ijk

-0.79

iants

-0.73

 Ruin

-0.68

ards

-0.68

eva

-0.65

wana

-0.65

ives

-0.64

anu

-0.64

arded

-0.64

artisan

-0.63

POSITIVE LOGITS

GROUND

0.83

quickShipAvailable

0.81

ttle

0.75

gren

0.75

mallow

0.72

 Holmes

0.72

fleet

0.71

croft

0.69

ldom

0.69

pine

0.68

Activations Density 0.053%

references to specific names, locations, and medical conditions

proper nouns and names associated with historical figures and locations

No Comments

No Known Activations