INDEX

Explanations

whitespace characters and coding comments

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

code structure keywords

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

estacks

-0.62

 ControllerBase

-0.61

NameInMap

-0.59

yntaxException

-0.58

-------

-0.58

 الرياضيه

-0.58

WithIdentifier

-0.56

 henvisninger

-0.54

>",

-0.54

 MotionEvent

-0.52

POSITIVE LOGITS

devamını

0.35

uxxxx

0.35

 nahilalakip

0.33

 Walkover

0.33

 cuales

0.32

 quedado

0.31

 Gött

0.29

FILL

0.29



0.29

הערות

0.29

Activations Density 0.011%

whitespace characters and coding comments

code structure keywords

No Comments

No Known Activations