INDEX

Explanations

references to specific items or actions that involve observing or discussing particular subjects

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

foreign or unusual words

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 تضيفلها

-0.63

 مرئيه

-0.61

 ModelExpression

-0.60

astéroïdes

-0.59

 resourceCulture

-0.58

RegressionTest

-0.56

KURZBESCHREIBUNG

-0.56

⅘

-0.52

RetentionPolicy

-0.51

usermodel

-0.50

POSITIVE LOGITS

こちらは

0.42

 тоже

0.38

grano

0.35

ywna

0.35

 Оп

0.34

too

0.33

=$?

0.33

fasst

0.33

 also

0.32

 bagay

0.31

Activations Density 0.077%

references to specific items or actions that involve observing or discussing particular subjects

foreign or unusual words

No Comments

No Known Activations