INDEX

Explanations

phrases related to functions, effectiveness, and interactions in various contexts

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

discussing actions or states

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

 MessageBoxIcon

-0.69

出版年

-0.67

 للمعارف

-0.62

 AssemblyTitle

-0.60

bootstrapcdn

-0.60

 autorytatywna

-0.59

 linkovi

-0.59

UnsafeEnabled

-0.58

 ſta

-0.57

classnames

-0.56

POSITIVE LOGITS

 işi

0.38

 perme

0.36

 froide

0.35

 fiduciary

0.34

 difficult

0.34

RegistryLite

0.33

 varying

0.33

暗い

0.33

atzen

0.32

activités

0.32

Activations Density 0.222%

phrases related to functions, effectiveness, and interactions in various contexts

discussing actions or states

No Comments

No Known Activations