INDEX

Explanations

specific formatting and punctuation patterns related to parentheses

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

Opening parentheses

np_acts-logits-general · gemini-2.0-flash

case insensitive "including"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 فيه

-0.45

forName

-0.44

 cóc

-0.44

 ocasião

-0.42

émoc

-0.42

 beryllium

-0.41

<bos>

-0.41

↵↵

-0.41

Representatives

-0.41

арак

-0.40

POSITIVE LOGITS

تقاوى

0.82

/=

0.70

 pinulongan

0.69

 including

0.66

というか

0.63

 INCLUDING

0.63

 Including

0.61

ie

0.60

 weihnachten

0.57

UBRE

0.57

Activations Density 0.326%

specific formatting and punctuation patterns related to parentheses

Opening parentheses

case insensitive "including"

No Comments

No Known Activations