INDEX

Explanations

occurrences of the name "Han" in various contexts

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

Following "Han" (Chinese/Hye/se/...)

np_acts-logits-general · gemini-2.0-flash

Han Chinese names

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

 lenker

-0.69

mergeFrom

-0.56

 SUDOC

-0.54

ContentAsync

-0.53

Bakgrunnsstoff

-0.53

Datuak

-0.52

########.

-0.51

pat

-0.51

 صوتيه

-0.50

 jspb

-0.50

POSITIVE LOGITS

ukkah

0.71

 condamné

0.63

ukah

0.59

 Solo

0.59

hydrogenase

0.57

 voudrais

0.57

Solo

0.56

]")]

0.55

 namanya

0.54

cestry

0.53

Activations Density 0.099%

occurrences of the name "Han" in various contexts

Following "Han" (Chinese/Hye/se/...)

Han Chinese names

No Comments

No Known Activations