INDEX

Explanations

proper nouns and fragments from programming code

oai_token-act-pair · gemini-2.0-flash

Location names

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 Svensson

-0.59

 original

-0.50

 global

-0.48

str

-0.47

Ju

-0.47

 real

-0.47

 Venti

-0.47

 grand

-0.46

 konk

-0.46

com

-0.46

POSITIVE LOGITS

UnsafeEnabled

1.03

 >=",

0.87

 Monfieur

0.85

 culoare

0.83

 ſtate

0.81

 tiegħek

0.81

 étoient

0.81

 anún

0.80

 parfüm

0.80

 BIBSYS

0.79

Activations Density 0.260%

proper nouns and fragments from programming code

Location names

No Comments

No Known Activations