INDEX

Explanations

mathematical notations and terms related to linear algebra and its applications

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Personensuche

-0.74

featureID

-0.73

+#+

-0.62

#+#

-0.60

 betweenstory

-0.59

 Walkover

-0.59

parsedMessage

-0.57

 privatisation

-0.57

EDEFAULT

-0.56

ReusableCell

-0.55

POSITIVE LOGITS

tagHelperRunner

0.38

 numbers

0.33

 område

0.33

 příslu

0.32

 grupp

0.32

 Grecs

0.32

 forskjellige

0.32

 Angaben

0.31

 diretamente

0.31

 number

0.31

Activations Density 0.047%

mathematical notations and terms related to linear algebra and its applications

No Comments

No Known Activations