INDEX

Explanations

references to law enforcement and military special operations units

oai_token-act-pair · gpt-3.5-turbo

references to special operations and teams in a military or law enforcement context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 contag

-0.74

 discont

-0.71

 breakup

-0.64

ynthesis

-0.63

ãĤ¨ãĥ«

-0.62

 inev

-0.61

ibal

-0.61

apest

-0.60

 decay

-0.59

istor

-0.58

POSITIVE LOGITS

Special

0.73

eers

0.71

 Squadron

0.67

rador

0.64

riors

0.63

ciating

0.62

èĢħ

0.62

ZE

0.61

 COURT

0.60

oise

0.60

Activations Density 0.139%

references to law enforcement and military special operations units

references to special operations and teams in a military or law enforcement context

No Comments

No Known Activations