INDEX

Explanations

words related to criminal activities, investigations, and law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to crime

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

heit

-0.69

 Centauri

-0.68

hetically

-0.67

ovo

-0.66

 Tennis

-0.65

chell

-0.65

arity

-0.65

 Airlines

-0.64

estine

-0.63

cher

-0.62

POSITIVE LOGITS

 spree

1.11

fighting

1.02

fighter

0.97

fighters

0.90

 scene

0.89

 ridden

0.86

 Scene

0.85

 prevention

0.85

 gangs

0.84

 perpetrated

0.83

Activations Density 0.039%

words related to criminal activities, investigations, and law enforcement

references to crime

No Comments

No Known Activations