INDEX

Explanations

terms related to ongoing processes or actions

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

with [state/action]

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

 чтобы

-0.46

 żeby

-0.42

出版年

-0.40

 quidem

-0.37

 щоб

-0.36

 AssemblyCompany

-0.35

 касается

-0.34

Recomendaciones

-0.34

 للمعارف

-0.34

 betweenstory

-0.34

POSITIVE LOGITS

MigrationBuilder

0.51

 يتيمه

0.50

 looming

0.46

 needing

0.44

 imminent

0.44

jsxFileName

0.43

protoimpl

0.42

OGND

0.42

 seemingly

0.42

 تضيفلها

0.42

Activations Density 0.013%

terms related to ongoing processes or actions

with [state/action]

No Comments

No Known Activations