INDEX

Explanations

numerical references and quantities related to measurement and statistics

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

out of

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

 Parkplatz

-0.34

 Übers

-0.33

 hundreds

-0.31

 Lebih

-0.31

 Zeitpunkt

-0.30

ándola

-0.30

kannya

-0.30

éndose

-0.30

 التأ

-0.30

late

-0.30

POSITIVE LOGITS

AndEndTag

0.75

 المعيارى

0.74



0.71

帖最后由

0.70

ImageContext

0.69

rungsseite

0.67

transQ

0.67

 AssemblyCompany

0.66

AddTagHelper

0.65

 initComponents

0.64

Activations Density 0.015%

numerical references and quantities related to measurement and statistics

out of

No Comments

No Known Activations