INDEX

Explanations

instances of people coming forward with information or accusations

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "come forward."

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

tein

-0.79

rip

-0.76

 Bots

-0.64

ľ

-0.63

icion

-0.63

Rocket

-0.62

OME

-0.62

Moj

-0.60

OL

-0.60

acca

-0.59

POSITIVE LOGITS

 testimonies

0.76

 voluntarily

0.76

 allegations

0.75

 toile

0.70

comings

0.70

wcs

0.68

osal

0.68

circle

0.66

 accusations

0.66

 submissions

0.61

Activations Density 0.013%

instances of people coming forward with information or accusations

instances of the phrase "come forward."

No Comments

No Known Activations