INDEX

Explanations

of

np_max-act · gemini-2.0-flash

This neuron activates on words that name actions or processes (i.e. nominalized verbs and gerunds such as “manipulation,” “allocation,” “instantiation,” “removal,” 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

up

-0.07

_None

-0.06

 Sullivan

-0.06

ecektir

-0.06

ign

-0.06

ид

-0.06

 कन

-0.06

-code

-0.06

ertura

-0.06

 convex

-0.06

POSITIVE LOGITS

 Antworten

0.08

InstanceOf

0.07

нием

0.06

 TEAM

0.06

着

0.06

kept

0.06

 outweigh

0.06

 Kanunu

0.06

２

0.06

子の

0.06

Activations Density 0.247%

of

This neuron activates on words that name actions or processes (i.e. nominalized verbs and gerunds such as “manipulation,” “allocation,” “instantiation,” “removal,” etc.).

No Comments

No Known Activations

of

This neuron activates on words that name actions or processes (i.e. nominalized verbs and gerunds such as “manipulation,” “allocation,” “instantiation,” “removal,” etc.).

No Comments

No Known Activations