INDEX

Explanations

study groups

np_max-act · gemini-2.0-flash

The neuron activates on numeric tokens—especially counts, measurements, and statistical figures (e.g. numbers and sample‐size annotations).

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

 larvae

-0.07

-Sep

-0.06

 funktion

-0.06

 đem

-0.06

	fflush

-0.06

 sanction

-0.06

険

-0.06

 guild

-0.06

 dong

-0.06

scrollView

-0.06

POSITIVE LOGITS

 countries

0.07

.Card

0.07

	BufferedReader

0.07

타이

0.07

иль

0.06

_RESOURCE

0.06

arım

0.06

Wes

0.06

χές

0.06

Activations Density 0.012%

study groups

The neuron activates on numeric tokens—especially counts, measurements, and statistical figures (e.g. numbers and sample‐size annotations).

No Comments

No Known Activations

study groups

The neuron activates on numeric tokens—especially counts, measurements, and statistical figures (e.g. numbers and sample‐size annotations).

No Comments

No Known Activations