INDEX

Explanations

numeric data and specific numeric references in a text

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

numbered years

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

 quantity

-0.41

ftagPool

-0.39

 quantità

-0.39

 محفوظة

-0.37

 piatta

-0.37

 quantidade

-0.36

 iniziale

-0.36

WithIdentifier

-0.36

 geha

-0.36

 personnelle

-0.35

POSITIVE LOGITS

 CanadaChoose

0.61

rungsseite

0.59

createCanvas

0.56

PyExc

0.51

Diwedd

0.48

 defaultstate

0.44

 ModelExpression

0.44

borderSide

0.44

EndProject

0.44

TypedDataSet

0.44

Activations Density 0.124%

numeric data and specific numeric references in a text

numbered years

No Comments

No Known Activations