INDEX

Explanations

occurrences of the word "money" in various forms

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

money market, money agreement

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 BoxDecoration

-0.51

 FormGroup

-0.43

borderSide

-0.39

출

-0.38

>--}}

-0.37

Alp

-0.36

 behalf

-0.35

 setObject

-0.35

futbolista

-0.35

tearDown

-0.35

POSITIVE LOGITS

Money

2.23

 Money

2.11

money

1.89

MONEY

1.81

 MONEY

1.77

 money

1.54

Geld

1.17

 Geld

1.09

 dinero

1.05

 dinheiro

0.96

Activations Density 0.008%

occurrences of the word "money" in various forms

money market, money agreement

No Comments

No Known Activations