INDEX

Explanations

user interactions and commands related to applications or software settings

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

items and lists

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

Externe

-0.37

 employed

-0.36

/**

-0.35

usado

-0.34

 eingesetzt

-0.34

 Spencer

-0.34

doi

-0.34

 Stevenson

-0.34

 yeterli

-0.33

➟

-0.33

POSITIVE LOGITS

NameInMap

0.69

 المعيارى

0.67

 propOrder

0.67

########.

0.66

AndEndTag

0.63

principalColumn

0.61

ConstraintMaker

0.59

 pinulongan

0.57

 disambiguazione

0.57

 intptr

0.57

Activations Density 0.031%

user interactions and commands related to applications or software settings

items and lists

No Comments

No Known Activations