INDEX

Explanations

elements that indicate clarity, coherence, and conversational tone in writing

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

words and phrases related to meaning, context, and intention in communication

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

tone and grammar

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

Билгалдахарш

-0.71

ciller

-0.58

adaptiveStyles

-0.58

 vPvB

-0.56

aarrggbb

-0.55

gewöhn

-0.55

*/;

-0.54

 newOwner

-0.54

Mediabestanden

-0.54

esehen

-0.54

POSITIVE LOGITS

 tone

1.26

 grammar

1.11

 grammatical

1.09

 Tone

0.97

Tone

0.95

 syntax

0.93

tone

0.93

 Grammar

0.91

EndContext

0.85

 formatting

0.83

Activations Density 0.204%

elements that indicate clarity, coherence, and conversational tone in writing

words and phrases related to meaning, context, and intention in communication

tone and grammar

No Comments

No Known Activations