INDEX

Explanations

repeated occurrences of the word "are" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

queryInterface

-1.02

rrggbb

-0.95

InjectAttribute

-0.83

styleType

-0.82

MLLoader

-0.73

RenderAtEndOf

-0.73

SequentialGroup

-0.73

 ContentValues

-0.71

lables

-0.71

NewGuid

-0.68

POSITIVE LOGITS

 enfans

0.68

 prêtres

0.60

 liné

0.55

 énergé

0.55

 yapı

0.54

ARE

0.53

 poches

0.52

wijze

0.52

exact

0.52

 térm

0.52

Activations Density 0.040%

repeated occurrences of the word "are" in various contexts

No Comments

No Known Activations