INDEX

Explanations

terms related to displays and imaging technology that do not require special accessories

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

using less

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

 bénévol

-0.34

 afectada

-0.33

 pravidel

-0.33

twimg

-0.33

 afectado

-0.33

 solidaridad

-0.33

ResponseData

-0.32

 diariamente

-0.32

 yaptı

-0.31

 kväll

-0.31

POSITIVE LOGITS

 ब्रेकडाउन

0.69

 NSCoder

0.65

 kaarangay

0.57

 MainAxisSize

0.53

 initComponents

0.53

 חיצוניים

0.53

 estekak

0.52

 doInBackground

0.52

providedIn

0.51

клопе

0.51

Activations Density 0.076%

terms related to displays and imaging technology that do not require special accessories

using less

No Comments

No Known Activations