INDEX

Explanations

references to accession numbers in scientific datasets

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

 accession

-1.55

 Accession

-1.34

Accession

-1.19

 accessions

-1.05

OGND

-0.67

 estekak

-0.64

 utafitiHapana

-0.63

StoryboardSegue

-0.62

oredCriteria

-0.56

NewUrlParser

-0.55

POSITIVE LOGITS

QRect

0.31

 مواليد

0.31

 cherchés

0.30

 Schmitz

0.28

ugd

0.28

ص

0.28

caire

0.28

ray

0.28

goal

0.28

校

0.28

Activations Density 0.010%

references to accession numbers in scientific datasets

No Comments

No Known Activations