INDEX

Explanations

nouns related to names, particularly in academic or literary contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.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

angue

-0.44

 CURIAM

-0.42

eip

-0.42

iestety

-0.41

hå

-0.41

folgung

-0.41

 fjor

-0.40

 finalmente

-0.39

joj

-0.39

 croce

-0.39

POSITIVE LOGITS

 متعلقه

0.89

jsxFileName

0.78

abestanden

0.77

 initComponents

0.75

Diweddarwch

0.74

fjspx

0.74

/−

0.68

 فريبيس

0.68

}/${

0.67

 ويكيپيديا

0.66

Activations Density 0.381%

nouns related to names, particularly in academic or literary contexts

No Comments

No Known Activations