INDEX

Explanations

references to personal experiences and emotions

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

personal context

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

ChildScrollView

-0.55

amaño

-0.50

 coolant

-0.50

 Signalez

-0.49

 Eats

-0.49

 tyres

-0.48

 lenker

-0.47

 waitFor

-0.47

avascript

-0.47

 válida

-0.47

POSITIVE LOGITS

 personal

1.14

Personal

1.08

 Personal

1.03

personal

1.02

 persönlichen

0.98

 persönliche

0.97

 personales

0.96

 persön

0.90

 PERSONAL

0.88

 pribadi

0.87

Activations Density 0.037%

references to personal experiences and emotions

personal context

No Comments

No Known Activations