INDEX

Explanations

various forms of numerical values and possibly identifiers in data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 NSCoder

-0.84

GEBURTSDATUM

-0.81

 ***!

-0.75

intios

-0.72

 ModelExpression

-0.69

<?

-0.69

">//

-0.66

 Roskov

-0.63

CodeAttribute

-0.63

valdo

-0.63

POSITIVE LOGITS

principalTable

0.59

 utafitiHapana

0.54

 areia

0.47

 مشين

0.44

 servire

0.42

נוּ

0.41

 כלל

0.41

 emel

0.40

pendenti

0.40

rantes

0.40

Activations Density 0.040%

various forms of numerical values and possibly identifiers in data

No Comments

No Known Activations