INDEX

Explanations

instances of individual letters or characters that may represent various elements or factors in structured data

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

single letters and formulas

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

 EconPapers

-0.81

 cherchés

-0.77

 linkovi

-0.75

 Italijanski

-0.75

TargetException

-0.70

featureID

-0.70

:+:

-0.69

rungsseite

-0.67

parsedMessage

-0.65

contentLoaded

-0.65

POSITIVE LOGITS

#%%

0.26

ONI

0.25

Per

0.25

 вот

0.25

 дека

0.23

 Achtung

0.23

Aud

0.22

ждён

0.22

bní

0.22

 custom

0.22

Activations Density 0.025%

instances of individual letters or characters that may represent various elements or factors in structured data

single letters and formulas

No Comments

No Known Activations