INDEX

Explanations

specific programming constructs and elements related to data structures and definitions

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

Rx, contact, tft, dirname, lat, row, int, pointer

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

GEBURTSDATUM

-0.72

expandindo

-0.71

 ब्रेकडाउन

-0.65

 становника

-0.63

 المعيارى

-0.62

transQ

-0.62

 estekak

-0.60

DBNull

-0.59

 Autorizaciones

-0.59

 AssemblyCulture

-0.59

POSITIVE LOGITS

 mutual

0.34

vy

0.33

 Angele

0.32

<h1>

0.32

ye

0.32

Ye

0.30

 absorbed

0.30

richts

0.30

ċi

0.30

ल

0.29

Activations Density 0.060%

specific programming constructs and elements related to data structures and definitions

Rx, contact, tft, dirname, lat, row, int, pointer

No Comments

No Known Activations