INDEX

Explanations

terms associated with rankings and identifiers

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

that end with "k"

np_acts-logits-general · gemini-2.0-flash

ending/containing k sound

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

lawayo

-0.80

volo

-0.78

帖最后由

-0.77

 EdgeInsets

-0.76

RectangleBorder

-0.74

 &___

-0.74

makeText

-0.72

providedIn

-0.69

Izvori

-0.65

uits

-0.64

POSITIVE LOGITS

kkkk

1.00

kkk

0.94

ⓧ

0.80

0.78

tsk

0.77

Duk

0.75

PKK

0.75

Cik

0.75

xk

0.75

KKKK

0.74

Activations Density 0.909%

terms associated with rankings and identifiers

that end with "k"

ending/containing k sound

No Comments

No Known Activations