INDEX

Explanations

technical writing

np_max-act · gemini-2.0-flash

This neuron fires on formal technical or domain-specific nouns (e.g. “providers,” “techniques,” “servers,” “analysis,” “policies”) typically found in patent-style or standards descriptions.

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

 clause

-0.06

 clauses

-0.06

 rein

-0.06

 lith

-0.06

ovie

-0.06

_keyword

-0.06

pad

-0.06

AL

-0.06

 encaps

-0.05

/en

-0.05

POSITIVE LOGITS

 convince

0.07

razil

0.07

бря

0.07

/red

0.07

及其

0.06

 France

0.06

/internal

0.06

-commerce

0.06

ibil

0.06

 Rocket

0.06

Activations Density 0.207%

technical writing

This neuron fires on formal technical or domain-specific nouns (e.g. “providers,” “techniques,” “servers,” “analysis,” “policies”) typically found in patent-style or standards descriptions.

No Comments

No Known Activations