INDEX

Explanations

references to the term "Han" and its variations across contexts

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

"Han" and following words

np_acts-logits-general · gemini-2.0-flash

Han Chinese

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

alty

-0.53

pat

-0.53

Bakgrunnsstoff

-0.51

 SUDOC

-0.51

ContentAsync

-0.49

 المعيارى

-0.47

 alternative

-0.47

kilde

-0.47

tsburgh

-0.46

Sil

-0.46

POSITIVE LOGITS

 magnétique

0.62

errHandler

0.61

 innehå

0.60

 vétérina

0.59

hydrogenase

0.58

 ziua

0.57

 strtok

0.57



0.57

 behövs

0.57

 dentaire

0.54

Activations Density 0.090%

references to the term "Han" and its variations across contexts

"Han" and following words

Han Chinese

No Comments

No Known Activations