INDEX

Explanations

technical specifications and numerical data

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

Diastolic

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

はじめに

-0.47

ftagPool

-0.39

nay

-0.36

bele

-0.36

 ExecuteAsync

-0.36

CodeAttribute

-0.36

紆

-0.35

DoubleQuotes

-0.35

之

-0.35

SourceChecksum

-0.35

POSITIVE LOGITS

NameInMap

0.52

 Normdatei

0.50

 ब्रेकडाउन

0.48

 שוליים

0.48

__(/*!

0.46

 tranſ

0.43

 verſ

0.42

 becauſe

0.42

 itſelf

0.42

 ſte

0.41

Activations Density 0.452%

technical specifications and numerical data

Diastolic

No Comments

No Known Activations