INDEX

Explanations

phrases related to food, dining, or restaurant recommendations

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

drivers, school, viruses

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 onAnimation

-0.49

 Réponses

-0.48

TargetException

-0.46

parsedMessage

-0.46

 AttributeSet

-0.45

rawDesc

-0.44

 Seeder

-0.43

 pulumi

-0.43

 GenerationType

-0.42

 ApiException

-0.42

POSITIVE LOGITS

pausal

0.50

ทั่ว

0.50

 cuisines

0.47

 algemene

0.47

 popolari

0.47

 Autorisations

0.46

 Económica

0.46

 ddelweddau

0.46

Vidite

0.45

 libraries

0.44

Activations Density 0.030%

phrases related to food, dining, or restaurant recommendations

drivers, school, viruses

No Comments

No Known Activations