INDEX

Explanations

Timepieces

np_max-act · gemini-2.0-flash

The neuron strongly activates on words naming time‐keeping elements (hours, minutes, seconds, dates, clock, time).

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

 Parameters

-0.06

 Ecology

-0.06

 yoktur

-0.06

 physiological

-0.06

 ecology

-0.06

 creek

-0.06

ulate

-0.06

 Incorpor

-0.06

 paradigm

-0.06

 personality

-0.06

POSITIVE LOGITS

ै।

0.06

戲

0.06

lah

0.06

RIA

0.06

كوم

0.06

Obama

0.06

[test

0.06

 ajud

0.06

[f

0.06

Activations Density 0.007%

Timepieces

The neuron strongly activates on words naming time‐keeping elements (hours, minutes, seconds, dates, clock, time).

No Comments

No Known Activations

Timepieces

The neuron strongly activates on words naming time‐keeping elements (hours, minutes, seconds, dates, clock, time).

No Comments

No Known Activations