INDEX

Explanations

references to locations and groups involved in social interactions

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

存于互联网档案馆

-0.75

RenderAtEndOf

-0.60

 سكانية

-0.58

 EconPapers

-0.55

 ligiloj

-0.54

 برانيه

-0.52

AccessorTable

-0.50

 transfieras

-0.50

 nahilalakip

-0.50

 Paglinawan

-0.50

POSITIVE LOGITS

intellectual

0.46

 educational

0.45

 intellectual

0.43

 Educational

0.42

 Education

0.41

providedIn

0.40

educational

0.40

 intelectual

0.38

 academic

0.38

Intellectual

0.37

Activations Density 0.121%

references to locations and groups involved in social interactions

No Comments

No Known Activations