INDEX

Explanations

verbose language, possibly from academic writing

oai_token-act-pair · gemini-2.0-flash

to

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

âtre

-0.48

Snoo

-0.47

новништво

-0.47

-0.46

 conscientes

-0.46

terno

-0.45

 genomen

-0.45

 capaces

-0.44

ovan

-0.44

 fiscales

-0.44

POSITIVE LOGITS

mybatisplus

0.88

 AssemblyProduct

0.72

tagHelperRunner

0.56

хьтан

0.55

Personensuche

0.53

Hentet

0.53

habhar

0.53

 EClass

0.53

|()

0.52

 Estatal

0.52

Activations Density 3.439%

verbose language, possibly from academic writing

to

No Comments

No Known Activations