INDEX

Explanations

specific formatting or markup elements in text, particularly related to data or code structures

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

amazon

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

Geplaatst

-0.94

WriteTagHelper

-0.61

Referencie

-0.54

Demografia

-0.54

/**

-0.54

WireFormat

-0.54

Abitanti

-0.53

 мәкалә

-0.51

 Signalez

-0.50

imageshack

-0.50

POSITIVE LOGITS

 AssemblyCulture

0.77

出版年

0.77

ⓧ

0.65

󠁿

0.64

msgTypes

0.60

tagHelper

0.57

tvguidetime

0.56

ndice

0.55

 Reverso

0.55

 ocio

0.54

Activations Density 0.006%

specific formatting or markup elements in text, particularly related to data or code structures

amazon

No Comments

No Known Activations