INDEX

Explanations

mentions or indications of an increase in quantity or intensity

oai_token-act-pair · gpt-3.5-turbo

instances of the word "increase" related to various contexts or phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Monstrous

-0.74

BuyableInstoreAndOnline

-0.72

 Healer

-0.68

 Voices

-0.68

 Impossible

-0.65

 Geneva

-0.64

feet

-0.64

 Surv

-0.63

 Bowman

-0.63

 Tycoon

-0.61

POSITIVE LOGITS

xual

1.24

 exponentially

0.90

 increment

0.85

ibal

0.85

 subsequ

0.83

 increases

0.82

uph

0.81

 increase

0.81

incre

0.81

baugh

0.80

Activations Density 0.029%

mentions or indications of an increase in quantity or intensity

instances of the word "increase" related to various contexts or phenomena

No Comments

No Known Activations