INDEX

Explanations

Software component names

np_max-act · gemini-2.0-flash

elements related to software licenses or legal terms.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

This neuron fires on programming identifiers—particularly names of classes, methods, and variables (i.e. code tokens like “Animator,” “installer,” “showInstaller,” etc.).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

пеки

-0.08

ственное

-0.07

 seas

-0.06

Sea

-0.06

 přítom

-0.06

 Peninsula

-0.06

 Beijing

-0.06

 Bitcoins

-0.06

 žádné

-0.06

POSITIVE LOGITS

er

0.16

ER

0.15

ers

0.13

aker

0.13

ator

0.12

ler

0.12

ker

0.11

ner

0.11

ter

0.11

izers

0.11

Activations Density 0.580%

Software component names

elements related to software licenses or legal terms.

This neuron fires on programming identifiers—particularly names of classes, methods, and variables (i.e. code tokens like “Animator,” “installer,” “showInstaller,” etc.).

No Comments

No Known Activations