INDEX

Explanations

hour

np_max-act · gemini-2.0-flash

The neuron selectively activates on floating-point number tokens (decimal fractions) in the text.

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

 Soil

-0.07

minimal

-0.06

(numero

-0.06

season

-0.06

 nevid

-0.06

 第三

-0.06

.getRoot

-0.06

/screen

-0.06

opcion

-0.06

already

-0.06

POSITIVE LOGITS

 nuestros

0.08

."""

0.07

 Palo

0.07

 تلك

0.06

 warrants

0.06

忠

0.06

 Lena

0.06

 grading

0.06

обще

0.06

 μέχρι

0.06

Activations Density 0.502%

hour

The neuron selectively activates on floating-point number tokens (decimal fractions) in the text.

No Comments

No Known Activations