INDEX

Explanations

emphasized descriptors of experiences or attributes related to continuous existence or belief

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

always been

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 للمعارف

-0.70

majánló

-0.67

 ddelwed

-0.65

aarrggbb

-0.62

 好文分享

-0.61

﹧

-0.60

 témoig

-0.60



-0.60

ロウィン

-0.59

󰀄

-0.59

POSITIVE LOGITS

 always

0.65

always

0.56

 Always

0.54

Always

0.52

 sempre

0.52

 never

0.47

 ALWAYS

0.46

昔から

0.45

 всегда

0.45

 traditionally

0.42

Activations Density 0.034%

emphasized descriptors of experiences or attributes related to continuous existence or belief

always been

No Comments

No Known Activations