INDEX

Explanations

phrases related to unexpected revelations or discoveries

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a revelation or unexpected outcome

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

è¦ļéĨĴ

-0.83

inately

-0.70

scape

-0.68

aph

-0.68

atro

-0.66

 Container

-0.66

agan

-0.65

antz

-0.65

mble

-0.64

necess

-0.64

POSITIVE LOGITS

terday

1.05

hift

0.75

 Turns

0.74

laus

0.74

turn

0.65

 turns

0.65

edit

0.64

ources

0.64

NetMessage

0.62

Initialized

0.62

Activations Density 0.011%

phrases related to unexpected revelations or discoveries

phrases indicating a revelation or unexpected outcome

No Comments

No Known Activations