INDEX

Explanations

phrases indicating a call to action

oai_token-act-pair · gpt-3.5-turbo

the use of imperative phrases and suggestions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

é¾įå¥ĳå£«

-0.73

ĪĴ

-0.71

 adolesc

-0.68

ĨĴ

-0.63

 depths

-0.62

 hormone

-0.62

ModLoader

-0.61

 autobi

-0.61

 traumatic

-0.60

 Depression

-0.59

POSITIVE LOGITS

encia

0.86

ndum

0.85

rison

0.81

unciation

0.78

ezvous

0.77

umbn

0.75

fecture

0.75

amaz

0.74

iners

0.72

inational

0.72

Activations Density 0.366%

phrases indicating a call to action

the use of imperative phrases and suggestions

No Known Activations