INDEX

Explanations

mentions of law enforcement officers in various contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of law enforcement personnel

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

FORE

-0.83

DAY

-0.82

Shock

-0.80

Tour

-0.74

izons

-0.70

Wind

-0.70

 Globe

-0.70

MED

-0.69

FACE

-0.68

WE

-0.68

POSITIVE LOGITS

 officers

1.17

 officer

1.06

 stationed

1.03

 Officers

0.97

 patrolling

0.91

paio

0.89

hip

0.84

arnaev

0.84

 detained

0.83

 academy

0.82

Activations Density 0.043%

mentions of law enforcement officers in various contexts

mentions of law enforcement personnel

No Comments

No Known Activations