INDEX

Explanations

references to checking and identifying servers, requests, or functionalities related to data management and programming contexts

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

determine, identify, or track

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

 autorytatywna

-0.91

 BoxDecoration

-0.84

 CanadaChoose

-0.83

 للاسماء

-0.79

AndEndTag

-0.78

Datuak

-0.73

 propOrder

-0.72

RectangleBorder

-0.70

исленность

-0.69

 صوتيه

-0.69

POSITIVE LOGITS

rõ

0.26

 last

0.26

Always

0.25

 Flücht

0.25

usammen

0.25

 Descub

0.24

 jelas

0.24

 perceber

0.24

lijkheid

0.24

 check

0.24

Activations Density 0.243%

references to checking and identifying servers, requests, or functionalities related to data management and programming contexts

determine, identify, or track

No Comments

No Known Activations