INDEX

Explanations

words related to crime and law enforcement

oai_token-act-pair · gpt-3.5-turbo

terms associated with crime or violations

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

upuncture

-0.77

ARA

-0.76

ascript

-0.75

fuck

-0.74

anship

-0.72

atism

-0.70

vet

-0.70

adan

-0.70

IDES

-0.68

orously

-0.68

POSITIVE LOGITS

 portion

0.82

 goods

0.75

 ones

0.70

 portions

0.69

escription

0.68

 Nost

0.67

 corpses

0.67

 corpse

0.67

 carc

0.66

ãĥ¤

0.65

Activations Density 0.232%

words related to crime and law enforcement

terms associated with crime or violations

No Comments

No Known Activations