INDEX

Explanations

HTML tags and elements

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

script tags and body

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

 otomatig

-0.84

 فريبيس

-0.81

 protoimpl

-0.78

enumii

-0.75

WriteBarrier

-0.72

تقاوى

-0.70

thrower

-0.68

bcryptjs

-0.68

原始内容存档于

-0.68

dataclass

-0.66

POSITIVE LOGITS

ViewFeatures

0.46

 huis

0.32

 main

0.32

 gobier

0.29

AutoScaleMode

0.28

 menu

0.28

ükü

0.27

 palme

0.27

фель

0.27

 campus

0.27

Activations Density 0.002%

HTML tags and elements

script tags and body

No Comments

No Known Activations