INDEX

Explanations

numerical values signifying important variables or outcomes in a mathematical context

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

math notation user

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

niſſe

-0.79

 CreateTagHelper

-0.77

ſſung

-0.77

 ddelwed

-0.76

iſchen

-0.73

 незавершена

-0.72

 témoig

-0.71

<unused41>

-0.71

[@BOS@]

-0.71

<unused3>

-0.71

POSITIVE LOGITS

AnimationsModule

0.33

 calef

0.31

or

0.28

 Vám

0.28

 pinulongan

0.27

 templo

0.27

 meurt

0.27

Hlav

0.27

 âme

0.27

0.26

Activations Density 0.000%

numerical values signifying important variables or outcomes in a mathematical context

math notation user

No Comments

No Known Activations