INDEX

Explanations

instances of graphical user interface elements or API features

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

names and dates

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

MENAFN

-0.47

 далее

-0.40

 maybe

-0.40

 Nieto

-0.40

居然

-0.39

maybe

-0.39

 lapin

-0.39

 Parece

-0.39

others

-0.38

 possibly

-0.37

POSITIVE LOGITS

 propOrder

0.80

setVerticalGroup

0.74

GEBURTSDATUM

0.70

Portail

0.69

 surla

0.69

Personensuche

0.66

GTCX

0.65

:✨

0.63

LabelTagHelper

0.63

LookAnd

0.61

Activations Density 0.001%

instances of graphical user interface elements or API features

names and dates

No Comments

No Known Activations