INDEX

Explanations

numerical indicators and parentheses related to references or citations

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

Enumerations or lists within parentheses

np_acts-logits-general · gemini-2.0-flash

(i), (ii), (iii), (1), (2), (3) followed by )

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.47

ay

-0.47

 infarction

-0.46

Camb

-0.45

 COPD

-0.43

Hello

-0.43

ilirsiniz

-0.43

getSingle

-0.43

-0.42

-0.41

POSITIVE LOGITS

WriteTagHelper

0.81

లాలు

0.77

下载附件

0.76

queryInterface

0.73

AnimationsModule

0.70

AddTagHelper

0.69

KommentareTeilen

0.69

بوابة

0.68

 оригіналу

0.66

AutoresizingMask

0.66

Activations Density 0.159%

numerical indicators and parentheses related to references or citations

Enumerations or lists within parentheses

(i), (ii), (iii), (1), (2), (3) followed by )

No Comments

No Known Activations