INDEX

Explanations

mentions of a specific individual named JP

oai_token-act-pair · gpt-3.5-turbo

references to JP Morgan

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

 thirds

-0.81

rums

-0.76

rights

-0.74

querque

-0.73

leneck

-0.71

mble

-0.67

staking

-0.66

mate

-0.66

lers

-0.63

thumbnails

-0.63

POSITIVE LOGITS

JP

1.17

oint

1.04

UNE

1.00

ACK

0.99

IME

0.97

NI

0.96

IT

0.95

EG

0.95

OY

0.94

ERG

0.93

Activations Density 0.010%

mentions of a specific individual named JP

references to JP Morgan

No Comments

No Known Activations