INDEX

Explanations

references to the term "Ki" and its variations, likely related to character names or identifiers in a specific context

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

Token "Ki" followed by another token

np_acts-logits-general · gemini-2.0-flash

Ki followed by various endings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

ніципа

-0.98

MENAFN

-0.80

 ComVisible

-0.74

XmlAccessorType

-0.73

LookAnd

-0.72

bufio

-0.70

endpush

-0.69

"");

-0.68

 OMIT

-0.67

 للمعارف

-0.67

POSITIVE LOGITS

///

0.43

:::

0.41

 iArr

0.40

 grieved

0.40

 blij

0.40

 profil

0.40

от

0.40

ting

0.39

ora

0.39

 дей

0.39

Activations Density 0.160%

references to the term "Ki" and its variations, likely related to character names or identifiers in a specific context

Token "Ki" followed by another token

Ki followed by various endings

No Comments

No Known Activations