INDEX

Explanations

proper nouns, especially names and titles

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.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

SharedDtor

-0.36

 naselje

-0.36

ponses

-0.34

 mémor

-0.34

 SIMBAD

-0.33

ntö

-0.33

vœ

-0.32

 succès

-0.32

.$,

-0.32

:✨

-0.32

POSITIVE LOGITS

oa̍t

0.63

saraba

0.62

tagHelperRunner

0.61

出版年

0.57

Gön

0.52

KURZBESCHREIBUNG

0.51

ddots

0.51

 wireType

0.50

 AssemblyTitle

0.49

 屋根

0.48

Activations Density 0.081%

proper nouns, especially names and titles

No Comments

No Known Activations