INDEX

Explanations

hexadecimal codes and values

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

0xff

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

__":

-0.75

adaptiveStyles

-0.66

__":

-0.66

modelBuilder

-0.65

دانشنامهٔ

-0.62

clero

-0.60

CreateModel

-0.59

thâu

-0.59

 فريبيس

-0.57

 useNavigate

-0.57

POSITIVE LOGITS

xFF

1.09

xff

0.99

ff

0.91

FF

0.84

FF

0.82

ff

0.82

iff

0.63

Ff

0.63

Ff

0.60

XFF

0.59

Activations Density 0.009%

hexadecimal codes and values

0xff

No Comments

No Known Activations