INDEX

Explanations

repeated instances of specific patterns or structures in text, particularly related to directional terms or categories

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

new energy, 20th-century, versus

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

 autorytatywna

-0.79

########.

-0.70

 مرئيه

-0.66

 disambiguazione

-0.65

 المعيارى

-0.61

webElementGuid

-0.60

Aholisi

-0.60

devamını

-0.58

 protoimpl

-0.56

httphttps

-0.56

POSITIVE LOGITS

findpost

0.37

AxisAlignment

0.35

cy

0.35

FailureListener

0.35

Ehrungen

0.33

IX

0.33

app

0.32

yes

0.32

pe

0.32

has

0.31

Activations Density 0.036%

repeated instances of specific patterns or structures in text, particularly related to directional terms or categories

new energy, 20th-century, versus

No Comments

No Known Activations