INDEX

Explanations

occurrences of the "<bos>" token, indicating the beginning of a new section or text

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

code and scientific terms

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

-0.68

 comercial

-0.52

in

-0.51

 high

-0.50

at

-0.50

 under

-0.50

 private

-0.49

-0.48

rasti

-0.48

 carrying

-0.48

POSITIVE LOGITS

 للمعارف

1.21

 betweenstory

1.13

Personensuche

1.05

 XCTest

1.01

 Мексичка

0.98

 MainAxisSize

0.97

djangoproject

0.96

 BorderRadius

0.96

ANTLR

0.95

 utafitiHapana

0.95

Activations Density 0.017%

occurrences of the "<bos>" token, indicating the beginning of a new section or text

code and scientific terms

No Comments

No Known Activations