INDEX

Explanations

mentions of incidents involving harm to people

oai_token-act-pair · gpt-3.5-turbo

references to casualties or numbers of people affected by violent events

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

soType

-0.84

inventoryQuantity

-0.78

soDeliveryDate

-0.76

é¾įåĸļå£«

-0.76

ventures

-0.75

taboola

-0.75

published

-0.75

FML

-0.73

å§«

-0.73

Management

-0.72

POSITIVE LOGITS

 unborn

0.83

 classmate

0.82

 innocent

0.82

 terrorists

0.82

 hostages

0.78

 intruder

0.78

 unarmed

0.75

 senseless

0.74

 rapist

0.73

 victims

0.71

Activations Density 0.261%

mentions of incidents involving harm to people

references to casualties or numbers of people affected by violent events

No Comments

No Known Activations