INDEX

Explanations

the term "OK" (optionally followed by a comma)

oai_token-act-pair · gemini-2.0-flash

OK

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

 Offisielt

-0.59

BeginContext

-0.59

 domicile

-0.56

 Horatio

-0.54

 defences

-0.48

Consent

-0.48

 noxious

-0.47

úsqueda

-0.47

 Fist

-0.47

 casualty

-0.46

POSITIVE LOGITS

IntoConstraints

0.70

mtrl

0.65

WriteTagHelper

0.63

AddTagHelper

0.62

 considérons

0.61

 localObject

0.59

"]];

0.57

?</

0.56

 fine

0.56

]}>

0.56

Activations Density 1.300%

the term "OK" (optionally followed by a comma)

OK

No Comments

No Known Activations