INDEX

Explanations

programming constructs related to object-oriented programming, such as class definitions and method calls

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

super calling parent methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nakalista

-0.51

OGND

-0.51

tagext

-0.48

RectangleBorder

-0.48

 Normdatei

-0.48

InSection

-0.47

ⓧ

-0.47

)++;

-0.46

uxxxx

-0.46

UnusedPrivate

-0.46

POSITIVE LOGITS

 tradisional

0.41

 normalen

0.40

 normales

0.39

 convencional

0.39

 gewöhn

0.38

通常の

0.38

 normalidad

0.37

 esprits

0.35

 recub

0.35

 souple

0.35

Activations Density 0.094%

programming constructs related to object-oriented programming, such as class definitions and method calls

super calling parent methods

No Comments

No Known Activations