INDEX

Explanations

contradictory or contrasting statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Hift

-0.77

 Anſ

-0.68

 myſelf

-0.64

 monks

-0.64

Efq

-0.64

MOC

-0.63

 Majefty

-0.63

twimg

-0.63

iedy

-0.62

$_"

-0.62

POSITIVE LOGITS

Datuak

0.62

0.54

awtextra

0.53

AxisAlignment

0.52

 StatefulWidget

0.51

it

0.51

 think

0.51

DataPropertyName

0.51

ichtliche

0.50

odeficiency

0.49

Activations Density 0.161%

contradictory or contrasting statements

No Comments

No Known Activations

contradictory or contrasting statements

No Comments

No Known Activations