INDEX

Explanations

phrases related to criminal activities, arrests, and charges

oai_token-act-pair · gpt-3.5-turbo

instances of arrests and related legal actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

casters

-0.70

emic

-0.69

 Remastered

-0.66

 impressions

-0.66

clips

-0.64

dust

-0.64

 superiority

-0.63

ankind

-0.62

rium

-0.61

BIT

-0.61

POSITIVE LOGITS

 criminally

0.83

 alongside

0.74

 handcuffed

0.70

osta

0.68

 arrested

0.67

 arra

0.67

by

0.66

arnaev

0.66

 administr

0.66

zai

0.66

Activations Density 0.093%

phrases related to criminal activities, arrests, and charges

instances of arrests and related legal actions

No Comments

No Known Activations