INDEX

Explanations

technology

np_max-act · gemini-2.0-flash

The neuron fires on named entities and other proper‐name tokens (e.g. titles, brand or program names).

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

_receive

-0.06

 Reality

-0.06

ze

-0.06

stride

-0.06

namespace

-0.06

ué

-0.06

 With

-0.06

，要

-0.06

ağın

-0.06

POSITIVE LOGITS

.Expr

0.07

广

0.06

}{↵

0.06

_texts

0.06

"|

0.06

ucz

0.06

 Coat

0.06

.subject

0.06

'}

0.06

toast

0.06

Activations Density 0.136%

technology

The neuron fires on named entities and other proper‐name tokens (e.g. titles, brand or program names).

No Comments

No Known Activations