INDEX

Explanations

formal terms and official statements or terminology

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

official and formal designations

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

Personensuche

-0.64

NameInMap

-0.50

CodeAttribute

-0.49

SharedCtor

-0.46

 HFILL

-0.45

 Italijani

-0.45

منابع

-0.44

opsida

-0.44

UserScript

-0.44

ScopeManager

-0.43

POSITIVE LOGITS

 offizielle

0.69

 officially

0.68

 offiziell

0.68

 official

0.66

 oficialmente

0.66

 oficjal

0.66

 resmi

0.65

 oficial

0.63

 officielle

0.60

 offiziellen

0.59

Activations Density 0.168%

formal terms and official statements or terminology

official and formal designations

No Comments

No Known Activations