INDEX

Explanations

instances of significant increases or changes in measurements or variables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

humble

-0.50

]<=

-0.50

häng

-0.47

 down

-0.47

 lowest

-0.47

minion

-0.46

 Michaels

-0.46

?')

-0.45

)<

-0.45

}<\

-0.44

POSITIVE LOGITS

ValueStyle

0.83

increase

0.75

 increase

0.72

 CanadaChoose

0.71

المناصب

0.71

Increases

0.71

⤴

0.70

Increase

0.68

 Paglinawan

0.68

">+

0.67

Activations Density 1.196%

instances of significant increases or changes in measurements or variables

No Comments

No Known Activations