INDEX

Explanations

Different languages

np_max-act · gemini-2.0-flash

The neuron is triggered by the Polish word “język” (and its token fragments) referring to “language.”

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

client

-0.08

(unit

-0.07

忙

-0.07

CHA

-0.06

Farm

-0.06

건

-0.06

 chan

-0.06

พยาบาล

-0.06

 Keith

-0.06

 ζω

-0.06

POSITIVE LOGITS

 sırasında

0.07

 Average

0.06

.RestController

0.06

nown

0.06

.SpringApplication

0.06

.RegularExpressions

0.06

(Base

0.06

 мови

0.06

.mybatis

0.06

	Output

0.06

Activations Density 0.022%

Different languages

The neuron is triggered by the Polish word “język” (and its token fragments) referring to “language.”

No Comments

No Known Activations

Different languages

The neuron is triggered by the Polish word “język” (and its token fragments) referring to “language.”

No Comments

No Known Activations