INDEX

Explanations

programming constructs related to event handling and signal processing

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

closing brackets or braces

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

ValueStyle

-0.83

setVerticalGroup

-0.81

XmlAccessorType

-0.78

 surla

-0.77

 ***!

-0.74

ftagPool

-0.73

 &___

-0.71

AddTagHelper

-0.69

 الرياضيه

-0.69

 MainAxisSize

-0.68

POSITIVE LOGITS

Noice

0.31

TokenNameCOMMA

0.31

 jää

0.30

 holiday

0.28

 Freizeit

0.28

 calles

0.28

ที

0.28

 voice

0.28

 đầu

0.27

 bahan

0.27

Activations Density 0.022%

programming constructs related to event handling and signal processing

closing brackets or braces

No Comments

No Known Activations