INDEX

Explanations

Hypothetical scenarios

np_max-act · gemini-2.0-flash

The neuron strongly activates on tokens that are parts of embedded numeric data—i.e. any tokens containing digits or decimal‐style numbers.

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

ではなく

-0.07

.track

-0.07

 Decode

-0.07

 grow

-0.06

_check

-0.06

در

-0.06

.unshift

-0.06

Linear

-0.06

LIST

-0.06

_pedido

-0.06

POSITIVE LOGITS

 HttpMethod

0.06

 Klopp

0.06

?>↵

0.06

------+

0.06

adr

0.06

 credentials

0.06

ﾝ

0.06

 أق

0.06

.SOCK

0.06

 accordion

0.06

Activations Density 0.063%

Hypothetical scenarios

The neuron strongly activates on tokens that are parts of embedded numeric data—i.e. any tokens containing digits or decimal‐style numbers.

No Comments

No Known Activations