INDEX

Explanations

mathematical constraints and equations

np_max-act · gemini-2.0-flash

This neuron activates on mathematical notation and equations (e.g. LaTeX-style formulas, Greek‐letter variables, integrals, derivatives).

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

stantiate

-0.06

basename

-0.06

 xlabel

-0.06

_tel

-0.06

coded

-0.06

 Means

-0.06

_SCHED

-0.06

	layout

-0.06

婆

-0.06

ti

-0.05

POSITIVE LOGITS

wsp

0.07

倍

0.07

 contentType

0.06

 buyer

0.06

cmp

0.06

 CODE

0.06

łu

0.06

 AREA

0.06

sou

0.06

 unas

0.06

Activations Density 0.046%

mathematical constraints and equations

This neuron activates on mathematical notation and equations (e.g. LaTeX-style formulas, Greek‐letter variables, integrals, derivatives).

No Comments

No Known Activations