INDEX

Explanations

th

np_max-act · gemini-2.0-flash

This neuron fires on numeric time markers—especially references to years, centuries, or decades (e.g., “20th,” “1920s,” etc.).

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

(Card

-0.07

рос

-0.07

رود

-0.07

-fed

-0.06

Jon

-0.06

 Dont

-0.06

 rebuilding

-0.06

Nickname

-0.06

 Does

-0.06

asyarak

-0.06

POSITIVE LOGITS

 twentieth

0.08

scrollTop

0.07

�

0.07

 Заг

0.07

.userid

0.07

 Civilization

0.06

^^^^

0.06

newInstance

0.06

.ExecuteNonQuery

0.06

 localObject

0.06

Activations Density 0.006%

th

This neuron fires on numeric time markers—especially references to years, centuries, or decades (e.g., “20th,” “1920s,” etc.).

No Comments

No Known Activations

th

This neuron fires on numeric time markers—especially references to years, centuries, or decades (e.g., “20th,” “1920s,” etc.).

No Comments

No Known Activations