INDEX

Explanations

phrases related to negative historical events or accusations

oai_token-act-pair · gpt-3.5-turbo

references to lynching or related themes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OTAL

-0.70

Republic

-0.69

OPLE

-0.67

Downloadha

-0.67

 Eternity

-0.65

quickShipAvailable

-0.64

avez

-0.63

displayText

-0.63

spring

-0.62

DERR

-0.62

POSITIVE LOGITS

lyn

1.15

sey

1.01

ching

0.96

seys

0.90

eteenth

0.86

zee

0.80

ched

0.78

poke

0.77

kees

0.75

nea

0.75

Activations Density 0.007%

phrases related to negative historical events or accusations

references to lynching or related themes

No Comments

No Known Activations