INDEX

Explanations

numeric data and statistical values

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

numbers and code snippets

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

 vuitton

-0.60

windowFixed

-0.60

dicionado

-0.58

 Numerade

-0.55

Datuak

-0.54

:+:

-0.54

 Италијани

-0.54

 betweenstory

-0.53

Espèce

-0.52

 disambiguazione

-0.52

POSITIVE LOGITS

GeoNames

0.39

 igual

0.38

UrlEncoded

0.37

 total

0.35

 simple

0.34

 demselben

0.33

 again

0.33

сылкі

0.33

 yine

0.33

同上

0.33

Activations Density 0.003%

numeric data and statistical values

numbers and code snippets

No Comments

No Known Activations