INDEX

Explanations

phrases related to police or law enforcement agencies

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals or entities associated with threats or legal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

witz

-0.84

igslist

-0.82

cv

-0.78

obyl

-0.74

zsche

-0.74

wcs

-0.71

hot

-0.71

hua

-0.70

amura

-0.69

story

-0.69

POSITIVE LOGITS

 Commissioner

1.06

 Officers

0.98

 Commissioners

0.95

 officers

0.94

 commissioner

0.91

 commissioners

0.90

 Inspect

0.87

 Administration

0.83

 officer

0.83

 investig

0.83

Activations Density 0.098%

phrases related to police or law enforcement agencies

references to specific individuals or entities associated with threats or legal issues

No Comments

No Known Activations