INDEX

Explanations

instances of "sense" followed by various descriptors, indicating a focus on perceptions or feelings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 Kidman

-0.81

könyv

-0.75

Optim

-0.73

 Alvarado

-0.73

 pestaña

-0.73

upaten

-0.73

 Barak

-0.72

AsUp

-0.71

 sério

-0.69

weds

-0.69

POSITIVE LOGITS

 SENSE

1.65

 Sense

1.52

 sense

1.40

Sense

1.34

sense

1.27

 senses

1.20

 Senses

1.02

 sensed

1.01

 senso

0.99

 Sensei

0.95

Activations Density 0.071%

instances of "sense" followed by various descriptors, indicating a focus on perceptions or feelings

No Comments

No Known Activations