INDEX

Explanations

sequences of characters or words that suggest a high activation or intensity, possibly indicating emphasis or importance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

endphp

-0.72

 utafitiHapana

-0.66

 szóci

-0.65

 estimés

-0.65

 WaitForSeconds

-0.61

 AuthService

-0.60

@}

-0.60

 Wallflower

-0.60

kering

-0.59

 تانيه

-0.59

POSITIVE LOGITS

MIDDLEWARE

0.68

 invokingState

0.64

ChildIndex

0.60

urlpatterns

0.56

0.55

routeProvider

0.53

rrggbb

0.51

she

0.51

 Piac

0.51

ácio

0.50

Activations Density 0.044%

sequences of characters or words that suggest a high activation or intensity, possibly indicating emphasis or importance

No Comments

No Known Activations

sequences of characters or words that suggest a high activation or intensity, possibly indicating emphasis or importance

No Comments

No Known Activations