INDEX

Explanations

highly informative and engaging content concerning guidance or reviews on specific topics

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

that start questions

np_acts-logits-general · gemini-2.0-flash

how, why, questions

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

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

 Roskov

-0.59

abestanden

-0.58

ؤلاء

-0.57

 للمعارف

-0.57

 AssemblyVersion

-0.56

SequentialGroup

-0.56

ReusableCell

-0.55

 ThemeData

-0.54

 transfieras

-0.53

ValueStyle

-0.53

POSITIVE LOGITS

how

1.54

How

1.45

How

1.44

how

1.35

why

1.22

Why

1.13

Why

1.11

如何

1.11

HOW

1.10

 cómo

1.07

Activations Density 0.188%

highly informative and engaging content concerning guidance or reviews on specific topics

that start questions

how, why, questions

No Comments

No Known Activations