INDEX

Explanations

components related to testing and application setup

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

running simulation, preclinical study

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

#+#

-0.52

tagHelperRunner

-0.49

 SwitchCompat

-0.47

astify

-0.46

HtmlAttribute

-0.46

delwed

-0.45

aarrggbb

-0.44

oplayer

-0.44

 Italijani

-0.44

+#+#

-0.44

POSITIVE LOGITS

 TestBed

0.54

AccessorTable

0.47

ſelf

0.42

Transkript

0.40

 nahilalakip

0.38

 vPvB

0.37

 geçi

0.36

 itſelf

0.35

StructEnd

0.35

 miniaturka

0.35

Activations Density 0.001%

components related to testing and application setup

running simulation, preclinical study

No Comments

No Known Activations