INDEX

Explanations

references to dates and numerical identifiers

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

identifiers and codes

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

Standard

-0.49

BibitemShut

-0.46

ſy

-0.45

AutoScaleMode

-0.45

ſelf

-0.44

 fubject

-0.43

setList

-0.43

ſelves

-0.43

RegressionTest

-0.43

Nueva

-0.42

POSITIVE LOGITS

ConstraintMaker

0.60

énario

0.52

tagHelperRunner

0.50

Демографія

0.45

الدراسه

0.44

 تضيفلها

0.43

GOTREF

0.42

AddTagHelper

0.41

Poloha

0.41

AddHtmlAttribute

0.41

Activations Density 0.001%

references to dates and numerical identifiers

identifiers and codes

No Comments

No Known Activations