INDEX

Explanations

learning

np_max-act · gemini-2.0-flash

The neuron activates on Portuguese terms related to artificial intelligence and machine learning (e.g. “inteligência,” “aprendizado,” “máquina,” “técnicas”).

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 interceptor

-0.06

outs

-0.06

 damit

-0.06

 scaled

-0.06

snake

-0.06

pan

-0.06

 Surely

-0.06

 olmadığını

-0.06

fung

-0.06

 между

-0.06

POSITIVE LOGITS

vn

0.07

(#)

0.07

 corrective

0.07

 getValue

0.07

 Moral

0.07

_Pr

0.06

 droit

0.06

규

0.06

Wo

0.06

 öff

0.06

Activations Density 0.011%

learning

The neuron activates on Portuguese terms related to artificial intelligence and machine learning (e.g. “inteligência,” “aprendizado,” “máquina,” “técnicas”).

No Comments

No Known Activations

learning

The neuron activates on Portuguese terms related to artificial intelligence and machine learning (e.g. “inteligência,” “aprendizado,” “máquina,” “técnicas”).

No Comments

No Known Activations