INDEX

Explanations

numeric values or codes, particularly in programming or technical contexts

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

hexadecimal strings and identifiers

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

 AssemblyCulture

-0.65

 nakalista

-0.59

 AssemblyVersion

-0.52

})->

-0.52

withIdentifier

-0.51

 Normdatei

-0.50

UnusedPrivate

-0.49

 للاسماء

-0.48

 permanently

-0.48

帖最后由

-0.48

POSITIVE LOGITS

kosh

0.39

Jereo

0.37

aneity

0.35

 illustratie

0.34

digheid

0.33

 Leverage

0.33

laştır

0.33

glycemia

0.33

 leverage

0.33

urture

0.33

Activations Density 0.011%

numeric values or codes, particularly in programming or technical contexts

hexadecimal strings and identifiers

No Comments

No Known Activations