INDEX

Explanations

phrases that signify achievement and reputation building

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

developed reputation or expertise

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

جغرافيا

-0.42

rimas

-0.36

 reinstall

-0.36

ur

-0.35

 disbursement

-0.35

Примітки

-0.34

pr

-0.34

 wasnt

-0.34

 Purposes

-0.34

Détails

-0.33

POSITIVE LOGITS

 earned

0.57

 amassed

0.56

 enviable

0.53

FieldBuilder

0.52

 reputation

0.52

 developed

0.52

 reputación

0.52

 للاسماء

0.52

 Earned

0.50

 reputations

0.49

Activations Density 0.026%

phrases that signify achievement and reputation building

developed reputation or expertise

No Comments

No Known Activations