INDEX

Explanations

phrases related to influencing or persuading others to do something

oai_token-act-pair · gpt-3.5-turbo

references to the action of persuading or getting others to do something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

anwhile

-0.83

ilial

-0.73

staking

-0.72

assetsadobe

-0.70

 intervals

-0.64

cember

-0.63

anges

-0.62

 margin

-0.62

 unexplained

-0.61

asionally

-0.61

POSITIVE LOGITS

orthy

0.72

 grips

0.67

Rid

0.67

FG

0.65

rid

0.65

plet

0.64

pleted

0.63

SIZE

0.63

enfranch

0.63

wake

0.63

Activations Density 0.149%

phrases related to influencing or persuading others to do something

references to the action of persuading or getting others to do something

No Comments

No Known Activations