INDEX

Explanations

/

np_max-act · gemini-2.0-flash

technical specifications and detailed instructions related to a subject or process.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

This neuron detects placeholder tokens (like “NAME_1”) used for templated list items or variable slots in the text.

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

 Cheese

-0.08

 Scotch

-0.07

 melt

-0.07

 cheese

-0.07

 submissive

-0.06

obs

-0.06

 consultations

-0.06

 мор

-0.06

 Blog

-0.06

 drilled

-0.06

POSITIVE LOGITS

_basis

0.08

_neighbor

0.06

(inode

0.06

ogonal

0.06

 Eğitim

0.06

 newY

0.06

rm

0.06

sł

0.06

osl

0.06

출

0.06

Activations Density 0.044%

/

technical specifications and detailed instructions related to a subject or process.

This neuron detects placeholder tokens (like “NAME_1”) used for templated list items or variable slots in the text.

No Comments

No Known Activations