INDEX

Explanations

features and functionalities related to applications or products

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

technical specifications

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

нгред

-0.82

 pinulongan

-0.79

擽

-0.72

 CanadaChoose

-0.71

iſchen

-0.71

 Wicidata

-0.71

 zwiſchen

-0.70

 unknownFields

-0.69

ſſung

-0.69

峫

-0.69

POSITIVE LOGITS

 Inscrivez

0.42

 devront

0.40

They

0.37

határoz

0.36

From

0.36

 faveur

0.33

zelfde

0.33

nimmt

0.33

 căr

0.33

According

0.32

Activations Density 0.002%

features and functionalities related to applications or products

technical specifications

No Comments

No Known Activations