INDEX

Explanations

references to sections or components in a document

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

The token "Part" followed by "of"

np_acts-logits-general · gemini-2.0-flash

part of

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

 تكبرها

-0.80

]--;

-0.71

Efq

-0.68

 Monfieur

-0.68

 autorità

-0.67

 purpoſe

-0.67

initializeApp

-0.67

 myſelf

-0.65

🏻‍♀️

-0.65

audiovisuel

-0.65

POSITIVE LOGITS

Parts

1.02

 Parts

0.98

 parts

0.94

 PARTS

0.91

parts

0.87

 PART

0.77

 partes

0.76

PART

0.72

cip

0.72

PARTS

0.70

Activations Density 0.104%

references to sections or components in a document

The token "Part" followed by "of"

part of

No Comments

No Known Activations