INDEX

Explanations

references to the name "Kim."

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

Kim (names or words)

np_acts-logits-general · gemini-2.0-flash

Kim followed by names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

Obrázky

-0.69

findpost

-0.68

DockStyle

-0.66

UnusedPrivate

-0.64

AutoScaleMode

-0.63

 MainAxisSize

-0.61

المناصب

-0.61

providedIn

-0.61

setopt

-0.60

prüche

-0.59

POSITIVE LOGITS

KIM

0.65

Kim

0.65

ber

0.60

KIM

0.59

تفصیلات

0.56

kim

0.53

 Jong

0.52

 Taehyung

0.52

balls

0.52

berley

0.52

Activations Density 0.096%

references to the name "Kim."

Kim (names or words)

Kim followed by names

No Comments

No Known Activations