INDEX

Explanations

terms and identifiers associated with programming or data structures

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

programming syntax and concepts

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

-0.50

 CascadeType

-0.50

principalColumn

-0.48

gsSP

-0.48

 beginnetje

-0.46

makedirs

-0.44

yntaxException

-0.44

oa̍t

-0.43

 Wikimedijinoj

-0.41

 المنام

-0.41

POSITIVE LOGITS

 ModelExpression

0.79

UnusedPrivate

0.59

XmlAccessType

0.57

 IBOutlet

0.43

 تضيفلها

0.38

nameof

0.37

erweise

0.36

 Outputs

0.36

webElement

0.36

 gepubliceerd

0.36

Activations Density 0.060%

terms and identifiers associated with programming or data structures

programming syntax and concepts

No Comments

No Known Activations