INDEX

Explanations

categorical and taxonomic classifications as well as identifiers in scientific descriptions

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

Taxonomic, universities, population

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

 betweenstory

-0.76

posedge

-0.54

RegressionTest

-0.51

rungsseite

-0.50

ConstraintMaker

-0.48

ChrTalk

-0.48

 CreateTagHelper

-0.47

CrossRef

-0.46

oredCriteria

-0.45

protoimpl

-0.45

POSITIVE LOGITS

 auroit

0.47

دانشنامهٔ

0.47

 kaarangay

0.45

rawDesc

0.43

 avoient

0.43

 feroit

0.42

 debout

0.41

 abstrato

0.41

 незавершена

0.40

 sauvages

0.39

Activations Density 0.000%

categorical and taxonomic classifications as well as identifiers in scientific descriptions

Taxonomic, universities, population

No Comments

No Known Activations