INDEX

Explanations

phrases indicating increases or additions

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

Hauptartikel

-0.69

">//

-0.58

 farlo

-0.58

deleteById

-0.57

 vitu

-0.56

inves

-0.55

 Futures

-0.55

OrWhiteSpace

-0.54

YMMV

-0.53

ularity

-0.52

POSITIVE LOGITS

 مواليد

0.63

 arric

0.63

KommentareTeilen

0.59

googleapis

0.57

>";

0.56

OPSIS

0.56

TagMode

0.56

owol

0.56

 administrativo

0.54

 initiale

0.52

Activations Density 0.135%

phrases indicating increases or additions

No Comments

No Known Activations