INDEX

Explanations

special formatting or markers in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 kasarigan

-0.73

pulumi

-0.68

Datuak

-0.67

########.

-0.65

 Baillargeon

-0.62

脚注の使い方

-0.60

 חיצוניים

-0.58

 Himo

-0.58

Географи

-0.57

 Carbide

-0.56

POSITIVE LOGITS

 suivants

0.64

ConstraintMaker

0.59

 suivantes

0.55

出版年

0.55

extAlignment

0.55

⮕

0.51

hdashline

0.50

BeginInit

0.50

SuspendLayout

0.48



0.48

Activations Density 0.041%

special formatting or markers in the text

No Comments

No Known Activations