INDEX

Explanations

structures and elements related to documents or data organization

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Code-related terms and category names

np_acts-logits-general · gemini-2.0-flash

legal or technical documentation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 الدولى

-0.69

</h1>

-0.57

…

-0.52

.*")]

-0.52

↵↵

-0.49

fsch

-0.47

↵↵↵

-0.47

</strong>

-0.47

<eos>

-0.45

prnewswire

-0.44

POSITIVE LOGITS

ArgsConstructor

0.98

 NDEBUG

0.92

ViewImports

0.87

Personendaten

0.86

 myſelf

0.79

ConstraintMaker

0.75

 Signalez

0.73

EndContext

0.73

 AppModule

0.72

〢

0.71

Activations Density 9.195%

structures and elements related to documents or data organization

Code-related terms and category names

legal or technical documentation

No Comments

No Known Activations