INDEX

Explanations

references to various crimes being committed, especially crimes against humanity and war crimes

oai_token-act-pair · gpt-3.5-turbo

references to crimes, particularly war crimes and crimes against humanity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

quickShipAvailable

-0.88

alg

-0.79

pole

-0.75

achev

-0.70

agles

-0.69

ulous

-0.69

flush

-0.64

agin

-0.63

ersed

-0.62

 snipp

-0.62

POSITIVE LOGITS

 committed

1.09

 perpetrated

1.04

 punishable

1.04

 spree

0.92

 crimes

0.87

 involving

0.84

 against

0.84

 offences

0.81

 Crimes

0.77

 offenses

0.75

Activations Density 0.032%

references to various crimes being committed, especially crimes against humanity and war crimes

references to crimes, particularly war crimes and crimes against humanity

No Comments

No Known Activations