INDEX

Explanations

numerical data and specific identifiers within technical or code-related contexts

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

Appears before digits

np_acts-logits-general · gemini-2.0-flash

numbers and version identifiers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

IsContent

-0.56

AddTagHelper

-0.50

cetines

-0.49

 menj

-0.48

oneofs

-0.48

lė

-0.47

TERMIN

-0.46

pædia

-0.46

pédie

-0.46

stood

-0.45

POSITIVE LOGITS

tvguidetime

0.50

transQ

0.44

 mCurrent

0.40

MigrationBuilder

0.40

 Мексичка

0.38

windowFixed

0.37

rivial

0.34

 tardi

0.34

 dataGridView

0.33

 BIBSYS

0.33

Activations Density 1.297%

numerical data and specific identifiers within technical or code-related contexts

Appears before digits

numbers and version identifiers

No Comments

No Known Activations