INDEX

Explanations

names or proper nouns related to notable individuals or entities

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

 Egipto

-0.47

 peinados

-0.46

PyErr

-0.43

 deuil

-0.43

appendChild

-0.41

 uñas

-0.41

 campañas

-0.40

 Perusahaan

-0.39

 Pferd

-0.39

 peinado

-0.39

POSITIVE LOGITS

GEBURTSDATUM

0.88

GEBURTS

0.77

MIR

0.69

GOTREF

0.68

 Auro

0.66

 ActionListener

0.64

 Omni

0.64

 Trusted

0.64

InjectAttribute

0.63

ORA

0.63

Activations Density 0.087%

names or proper nouns related to notable individuals or entities

No Comments

No Known Activations