INDEX

Explanations

structured components or sections typically found in documentation or code comments

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

code comments

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

 صوتيه

-0.71

fjspx

-0.68

__':

-0.63

 surla

-0.63

 betweenstory

-0.62

oredCriteria

-0.60

RegressionTest

-0.58

 otomatig

-0.58

localctx

-0.57

 TestBed

-0.57

POSITIVE LOGITS

0.32

background

0.31

protoimpl

0.30

 bParam

0.29

complexContent

0.28

 anteriore

0.27

 opgenomen

0.27

+:+

0.26

 expanding

0.25

Activations Density 0.037%

structured components or sections typically found in documentation or code comments

code comments

No Comments

No Known Activations