INDEX

Explanations

references to the word "Eagle"

oai_token-act-pair · gpt-3.5-turbo

references to "Eagle."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ERN

-0.66

*/(

-0.66

ÑĮ

-0.65

UME

-0.64

 Rowling

-0.64

 Lanka

-0.63

akuya

-0.62

esters

-0.62

 nostalg

-0.61

ĸļ

-0.61

POSITIVE LOGITS

 Eagle

1.08

 Scout

0.93

horn

0.90

agle

0.87

agles

0.82

 Feather

0.80

zeb

0.79

 Claw

0.77

wings

0.75

imore

0.75

Activations Density 0.008%

references to the word "Eagle"

references to "Eagle."

No Comments

No Known Activations