INDEX

Explanations

references to individuals, particularly names and their associated attributes or roles

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IntoConstraints

-0.50

قایناق‌لار

-0.48

zzleHttp

-0.48

 resourceCulture

-0.47

 Commencez

-0.45

WebElementEntity

-0.45

MockBean

-0.44

 transfieras

-0.42

 للمعارف

-0.42

+#+#

-0.40

POSITIVE LOGITS

AndEndTag

0.54

 retaliation

0.51

Uwagi

0.45

Counter

0.40

isAuthenticated

0.40

bert

0.40

 دادن

0.39

ded

0.39

krom

0.39

 Counter

0.39

Activations Density 0.015%

references to individuals, particularly names and their associated attributes or roles

No Comments

No Known Activations