INDEX

Explanations

references to Osama bin Laden

oai_token-act-pair · gpt-3.5-turbo

mentions of Osama bin Laden

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

GRE

-0.72

 Constantin

-0.72

ISS

-0.70

Wide

-0.70

PASS

-0.69

Premium

-0.66

CAP

-0.66

ALT

-0.64

Medic

-0.63

WD

-0.63

POSITIVE LOGITS

 Laden

1.13

abad

0.94

querque

0.87

gebra

0.86

 Jinping

0.84

addafi

0.83

bin

0.79

selage

0.79

ghan

0.79

enstein

0.79

Activations Density 0.012%

references to Osama bin Laden

mentions of Osama bin Laden

No Comments

No Known Activations