INDEX

Explanations

code comments and labels in a programming context

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

Code, function calls, or programming syntax

np_acts-logits-general · gemini-2.0-flash

判断卡牌

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

twimg

-0.58

 feroit

-0.57

OGND

-0.56

 vooz

-0.56

ロウィン

-0.56

ſammen

-0.55

 témoig

-0.54

 تضيفلها

-0.53

 auroit

-0.52

 stället

-0.52

POSITIVE LOGITS

uxxxx

0.63

دانشنامهٔ

0.47

حياته

0.46

transQ

0.44

 محفوظة

0.42

ⓧ

0.39

ueras

0.37

 незавершена

0.37

nodoc

0.37

 урна

0.36

Activations Density 0.056%

code comments and labels in a programming context

Code, function calls, or programming syntax

判断卡牌

No Comments

No Known Activations