INDEX

Explanations

names of researchers or authors in scientific 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

Aiheesta

-0.60

 samband

-0.57

 bán

-0.53

})*/

-0.53

EVA

-0.52

 gefangen

-0.51

ไซ

-0.51

utives

-0.51

marshaller

-0.50

떻

-0.50

POSITIVE LOGITS

PMailer

0.81

EndInit

0.65

IsMutable

0.60

esModule

0.58

__':

0.53

posedge

0.53

weds

0.52

ution

0.52

 ostr

0.51

pozorn

0.50

Activations Density 0.471%

names of researchers or authors in scientific contexts

No Comments

No Known Activations