INDEX

Explanations

terms related to movement or progress

oai_token-act-pair · gpt-3.5-turbo

instances of the word "go" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

lihood

-0.75

lished

-0.73

 Advertisement

-0.71

alia

-0.69

stakes

-0.66

 Responsibility

-0.64

expression

-0.62

andestine

-0.61

vich

-0.60

majority

-0.59

POSITIVE LOGITS

go

2.97

Go

1.83

go

1.81

Go

1.75

 goes

1.61

GO

1.56

 proceed

1.42

 went

1.38

 gone

1.36

 stay

1.25

Activations Density 0.045%

terms related to movement or progress

instances of the word "go" in various contexts

No Comments

No Known Activations