INDEX

Explanations

numerical data related to quantities or identifiers

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

dates and numbers

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

帖最后由

-0.74

 createSprite

-0.57

 فريبيس

-0.55

 '\\;'

-0.54

ScopeManager

-0.51

цездатний

-0.49

RenderAtEndOf

-0.49

principalColumn

-0.49

GEBURTSDATUM

-0.47

 nahilalakip

-0.46

POSITIVE LOGITS

 ২০১

0.55

 competencia

0.50

 urgencia

0.46

 deseado

0.46

 verão

0.45

loß

0.45

loga

0.45

 Grun

0.44

 ۲۰۱

0.44

 failings

0.43

Activations Density 0.007%

numerical data related to quantities or identifiers

dates and numbers

No Comments

No Known Activations