INDEX

Explanations

(

np_max-act · gemini-2.0-flash

The neuron is detecting date‐ and year‐related tokens—especially multi‐digit numerals and parenthetical birth/death years—in timeline or historical-entry contexts.

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

(form

-0.07

/task

-0.07

ها

-0.07

 resolved

-0.06

<nav

-0.06

COVERY

-0.06

 Justice

-0.06

bound

-0.06

UAL

-0.06

łe

-0.06

POSITIVE LOGITS

Weight

0.07

_tid

0.07

iß

0.06

_ctxt

0.06

 repaint

0.06

 Δι

0.06

ěst

0.06

_quad

0.06

 siyas

0.06

 verr

0.06

Activations Density 0.003%

(

The neuron is detecting date‐ and year‐related tokens—especially multi‐digit numerals and parenthetical birth/death years—in timeline or historical-entry contexts.

No Comments

No Known Activations

(

The neuron is detecting date‐ and year‐related tokens—especially multi‐digit numerals and parenthetical birth/death years—in timeline or historical-entry contexts.

No Comments

No Known Activations