INDEX

Explanations

phrases related to actions and social interactions

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

every time

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

 propOrder

-0.68

contentLoaded

-0.51

fxml

-0.45

CreateModel

-0.43

 Wiese

-0.42

 onely

-0.42

classnames

-0.40

 Houſe

-0.40

__]

-0.39

 schal

-0.38

POSITIVE LOGITS

Whenever

0.72

 Whenever

0.71

whenever

0.69

 whenever

0.65

 everytime

0.56

每一次

0.54

Anytime

0.48

たびに

0.47

 resourceCulture

0.46

tgär

0.45

Activations Density 0.028%

phrases related to actions and social interactions

every time

No Comments

No Known Activations