INDEX

Explanations

period

np_max-act · gemini-2.0-flash

The neuron activates on numeric timestamp and version‐number sequences (e.g. date/time stamps and dotted version strings) in diff/file‐header metadata.

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

(Dictionary

-0.06

methods

-0.06

.mapper

-0.06

enders

-0.06

 anyhow

-0.06

(Collectors

-0.06

 함수

-0.06

.dataset

-0.06

OUNTER

-0.06

.Popup

-0.06

POSITIVE LOGITS

 oblivious

0.07

 Josef

0.07

()}}↵

0.07

 geçir

0.07

eid

0.07

 Kingston

0.07

 ubic

0.07

 setId

0.07

 Jude

0.07

 anesthesia

0.06

Activations Density 0.002%

period

The neuron activates on numeric timestamp and version‐number sequences (e.g. date/time stamps and dotted version strings) in diff/file‐header metadata.

No Comments

No Known Activations