INDEX

Explanations

phrases related to incentives or motivators

oai_token-act-pair · gpt-3.5-turbo

references to incentives in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

rooms

-0.94

gaard

-0.79

room

-0.76

bane

-0.75

angers

-0.74

abad

-0.73

 Wand

-0.72

thing

-0.71

agn

-0.70

geist

-0.69

POSITIVE LOGITS

 incentive

1.15

 incentives

1.08

 incentiv

1.06

 incent

1.03

 compensation

0.87

 schemes

0.87

 rewarded

0.86

 compulsion

0.86

Reviewer

0.85

 compel

0.85

Activations Density 0.016%

phrases related to incentives or motivators

references to incentives in various contexts

No Comments

No Known Activations