INDEX

Explanations

function and method calls related to clearing data or memory structures

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

clearing/emptying content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

HostException

-0.31

 enmar

-0.28

arned

-0.26

 defective

-0.26

 Defective

-0.26

mær

-0.26

 gelieb

-0.25

komsten

-0.24

 eterno

-0.24

tangentMode

-0.24

POSITIVE LOGITS

 clearing

1.96

 cleared

1.84

 cleansing

1.77

 clears

1.77

 clean

1.77

 cleans

1.75

 cleaning

1.73

clearing

1.73

 Clearing

1.70

clean

1.69

Activations Density 0.217%

function and method calls related to clearing data or memory structures

clearing/emptying content

No Comments

No Known Activations