INDEX

Explanations

phrases related to advertisements

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Ad" indicating advertising-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 metab

-0.66

 DRAG

-0.65

Zur

-0.64

 behav

-0.64

 tomat

-0.64

 blah

-0.62

 Sting

-0.61

Fn

-0.61

 Tacoma

-0.61

 ÏĦ

-0.60

POSITIVE LOGITS

vertising

1.41

roximately

1.35

elaide

1.21

elligence

1.16

resa

1.16

withstanding

1.14

ellectual

1.13

cluding

1.11

ropolitan

1.10

xon

1.10

Activations Density 0.125%

phrases related to advertisements

instances of the word "Ad" indicating advertising-related content

No Comments

No Known Activations