INDEX

Explanations

phrases related to discussions about media, events, or decision-making processes

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

focuses on

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

 <<<<<<<<<<<<<<

-0.65

OGND

-0.64

sizeCache

-0.61

 estekak

-0.61

 EconPapers

-0.54

RegressionTest

-0.52

Enllaços

-0.50

Rujuakan

-0.49

.*")]

-0.48

RegistryLite

-0.47

POSITIVE LOGITS

posedge

0.39

ंदीखरीदारी

0.37

 Notepad

0.37

 Assessor

0.37

PCell

0.37

menubar

0.36

bitField

0.36

blurRadius

0.36

flashdata

0.35

 holocaust

0.35

Activations Density 0.020%

phrases related to discussions about media, events, or decision-making processes

focuses on

No Comments

No Known Activations