INDEX

Explanations

fragmented, descriptive sentences, especially those using the word "a".

oai_token-act-pair · gemini-2.0-flash

a

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

honneur

-0.59

expandindo

-0.59

twimg

-0.57

Personensuche

-0.57

Rhestr

-0.56

 Administrativna

-0.56

DeleteBehavior

-0.56

 Wikispecies

-0.54

tonode

-0.51

LookAnd

-0.51

POSITIVE LOGITS

 ActionBar

0.54

aien

0.49

 Argos

0.47

Apis

0.47

 Amaz

0.46

Abi

0.46

Ep

0.46

 Antig

0.45

 Attention

0.44

 Ayuda

0.44

Activations Density 2.680%

fragmented, descriptive sentences, especially those using the word "a".

a

No Comments

No Known Activations