INDEX

Explanations

tokens or identifiers related to programming and data structures

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

 فريبيس

-0.80

 '\\;'

-0.78

tagHelperRunner

-0.73

 disambiguazione

-0.68

ScopeManager

-0.66

 <>",

-0.66

writeFieldEnd

-0.63

adpleegd

-0.62

queryInterface

-0.60

NameInMap

-0.59

POSITIVE LOGITS

modified

0.36

cupertino

0.34

preview

0.34

 enfans

0.33

зив

0.33

 Modified

0.32

brother

0.31

rmdir

0.31

 Help

0.31

ez

0.31

Activations Density 0.065%

tokens or identifiers related to programming and data structures

No Comments

No Known Activations