INDEX

Explanations

aggregation functions

np_max-act · gemini-2.0-flash

This neuron detects functional “fold‐style” aggregation calls (e.g. reduceLeft/foldLeft/accumulate/inject) over sequences.

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

 cropping

-0.07

.hex

-0.07

(Constructor

-0.07

_wallet

-0.07

^{[

-0.06

 excluding

-0.06

 window

-0.06

(delegate

-0.06

.csv

-0.06

_given

-0.06

POSITIVE LOGITS

vir

0.07

ILLED

0.06

 còn

0.06

現

0.06

 getColumn

0.06

 كسارة

0.06

!'

0.06

 аль

0.06

 volte

0.06

�

0.06

Activations Density 0.027%

aggregation functions

This neuron detects functional “fold‐style” aggregation calls (e.g. reduceLeft/foldLeft/accumulate/inject) over sequences.

No Comments

No Known Activations