INDEX

Explanations

configuration-related terms and elements within a programming context

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

ConstraintMaker

-0.48

itinéraire

-0.48

Autoritní

-0.46

AndEndTag

-0.45

EndInit

-0.42

 utafitiHapana

-0.41

PyExc

-0.40

htbp

-0.40

Portail

-0.39

asList

-0.39

POSITIVE LOGITS

:✨

0.72

transQ

0.49

cic

0.42

EndGlobalSection

0.41

cio

0.39

 restri

0.38

 ModelExpression

0.38

 potr

0.38

 Meksiku

0.37

 fran

0.37

Activations Density 0.048%

configuration-related terms and elements within a programming context

No Comments

No Known Activations