INDEX

Explanations

phrases related to the name "Brian."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the character combination "Br" 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 @ 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

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.82

 AMERICA

-0.78

eers

-0.75

WARE

-0.75

PLIED

-0.74

ãģŁ

-0.66

heid

-0.66

ei

-0.66

æĸ¹

-0.65

YA

-0.65

POSITIVE LOGITS

acket

1.27

anches

1.23

atton

1.16

unn

1.13

ackets

1.12

anch

1.09

idget

1.08

inger

1.07

anson

1.07

anca

1.06

Activations Density 0.013%

phrases related to the name "Brian."

occurrences of the character combination "Br" in various contexts

No Comments

No Known Activations