INDEX

Explanations

references to programming constructs and database queries

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

awtextra

-0.50

jsii

-0.49

 AttributeSet

-0.48

хьтан

-0.48

 kasarigan

-0.47

tagHelperRunner

-0.47

 NSCoder

-0.45

Comté

-0.43

DockStyle

-0.43

chartInstance

-0.42

POSITIVE LOGITS

 للمعارف

0.50

łyż

0.43

 kılıf

0.42

droje

0.41

 baskılı

0.40

RTSN

0.40

 küpe

0.40

 fontes

0.40

 سكانية

0.40

 bolsillo

0.38

Activations Density 0.090%

references to programming constructs and database queries

No Comments

No Known Activations