INDEX

Explanations

sections that contain numerical data or references to figures

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

before or related to a factor

np_acts-logits-general · gemini-2.0-flash

cannot rely

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 lettera

-0.50

 grecque

-0.48

 éducation

-0.47

)/

-0.47

 soigne

-0.47

 koska

-0.46

 chaîne

-0.46

 distritos

-0.46

 lycée

-0.46

 لأن

-0.45

POSITIVE LOGITS

PerformLayout

0.86

脚注の使い方

0.84

 ExecuteAsync

0.81

 Thursday

0.72

ChildScrollView

0.71

 AttributeSet

0.71

 Saturday

0.70

 ostavi

0.69

 invokingState

0.67

 itſelf

0.67

Activations Density 0.215%

sections that contain numerical data or references to figures

before or related to a factor

cannot rely

No Comments

No Known Activations