INDEX

Explanations

numeric data and structural elements in a document

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

denote or represent

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

CROSSTALK

-0.42

Становништво

-0.39

WebElementEntity

-0.38

原始内容存档于

-0.37

 InputDecoration

-0.36

 nakalista

-0.35

 palju

-0.33

msgTypes

-0.32

pture

-0.32

SaveVideo

-0.32

POSITIVE LOGITS

 belonged

0.56

 belongs

0.54

eleste

0.49

InitStruct

0.48

esez

0.47

 owes

0.47

 belong

0.47

 optional

0.47

 corresponds

0.47

 represents

0.46

Activations Density 0.155%

numeric data and structural elements in a document

denote or represent

No Comments

No Known Activations