INDEX

Explanations

references to various forms of media, particularly focusing on auditory, visual, and written content

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

research methods and data collection

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

 ExecuteAsync

-0.45

styleable

-0.42

MLLoader

-0.42

RTSC

-0.41

LEADING

-0.41

 spec

-0.41

 material

-0.40

 Denis

-0.39

rh

-0.39

lease

-0.39

POSITIVE LOGITS

versies

0.50

culada

0.46

OGND

0.45

 nią

0.44

irmanship

0.43

GOTREF

0.43

 enfermed

0.42

 überrascht

0.41

MessageTagHelper

0.41

 ciudadanía

0.41

Activations Density 0.045%

references to various forms of media, particularly focusing on auditory, visual, and written content

research methods and data collection

No Comments

No Known Activations