INDEX

Explanations

the phrase "integral part" or the words "in whole or in part"

oai_token-act-pair · gemini-2.0-flash

technical terms related to integration

np_token-act-pair-logits · gpt-4o-mini

say "part"

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 components

-1.53

 parts

-1.44

components

-1.41

 elements

-1.38

Components

-1.38

parts

-1.33

Parts

-1.32

 Components

-1.31

 Parts

-1.30

 component

-1.27

POSITIVE LOGITS

 ported

0.63

 EClass

0.59

windowFixed

0.58

portál

0.56

 port

0.54

kháu

0.53

getPort

0.52

 Certo

0.52

obenz

0.50

depre

0.50

Activations Density 0.979%

the phrase "integral part" or the words "in whole or in part"

technical terms related to integration

say "part"

No Comments

No Known Activations