INDEX

Explanations

mentions of a person evading or dealing with situations involving conflict or consequences

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

)=(

-0.73

 Methods

-0.67

Reviewer

-0.66

 Dispatch

-0.64

 Handling

-0.64

ONSORED

-0.63

 Prosecutor

-0.61

rification

-0.60

DOC

-0.59

mus

-0.59

POSITIVE LOGITS

omsday

1.37

ppel

1.29

herty

1.25

gging

1.09

lez

1.07

ctors

1.05

ctr

1.04

berman

1.03

ozy

1.02

ggie

1.01

Activations Density 0.048%

mentions of a person evading or dealing with situations involving conflict or consequences

No Comments

No Known Activations