INDEX

Explanations

titles and ranks within law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement personnel and their titles

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

DragonMagazine

-0.75

OPA

-0.72

era

-0.72

ahime

-0.67

SPONSORED

-0.66

pless

-0.66

worn

-0.65

dollar

-0.65

 panc

-0.65

-0.64

POSITIVE LOGITS

Lt

0.96

 Investigator

0.92

 Inspector

0.92

 Joseph

0.86

Maj

0.85

 Superintendent

0.84

geant

0.83

 Deputy

0.83

 Chief

0.81

 officer

0.80

Activations Density 0.200%

titles and ranks within law enforcement

references to law enforcement personnel and their titles

No Comments

No Known Activations