INDEX

Explanations

proper nouns, particularly names of people

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

 gray

-0.40

isdir

-0.40

 different

-0.39

دانشنامهٔ

-0.38

writeFieldEnd

-0.37

rinks

-0.37

 grey

-0.37

 ways

-0.37

 local

-0.36

 fruits

-0.36

POSITIVE LOGITS

 AssemblyCulture

0.52

 Paglinawan

0.51

PreferredItem

0.50

celotti

0.50

########.

0.48

 صوتيه

0.48

Insee

0.47

 himself

0.47

 stesso

0.47

 transfieras

0.46

Activations Density 0.037%

proper nouns, particularly names of people

No Comments

No Known Activations