INDEX

Explanations

phrases indicating availability and reference to known items or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 autorytatywna

-0.76

parsedMessage

-0.73

Jeografia

-0.70

 AssemblyVersion

-0.69

Izvori

-0.66

OGND

-0.65

 intStringLen

-0.62

 ujednoznacz

-0.60

utilisons

-0.58

genodigd

-0.57

POSITIVE LOGITS

also

0.59

Also

0.57

 also

0.53

也非常

0.52

 également

0.52

 myös

0.50

 additionally

0.50

 Также

0.50

Также

0.50

 Also

0.49

Activations Density 0.195%

phrases indicating availability and reference to known items or concepts

No Comments

No Known Activations