INDEX

Explanations

references to happiness or related emotional states

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PathVariable

-0.69

र्भ

-0.68

 zagran

-0.67

 tehd

-0.66

 Borges

-0.66

 Sten

-0.65

 мной

-0.65

Sten

-0.65

hithe

-0.64

 conducts

-0.64

POSITIVE LOGITS

 happy

1.47

 Happy

1.42

 HAPPY

1.42

HAPPY

1.41

 happiness

1.39

 Happiness

1.35

happy

1.27

 happier

1.27

 HAPP

1.26

happiness

1.26

Activations Density 0.046%

references to happiness or related emotional states

No Comments

No Known Activations