INDEX

Explanations

EST

np_max-act · gemini-2.0-flash

This neuron primarily lights up on common short function words—prepositions and articles—rather than on substantive content terms.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

references to statistical data and probabilities related to frequency and averages.

oai_token-act-pair · gpt-4o-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

 legitim

-0.07

маз

-0.07

.mm

-0.07

 personalize

-0.07

ception

-0.07

(strcmp

-0.06

 pagar

-0.06

functional

-0.06

ฬ

-0.06

픈

-0.06

POSITIVE LOGITS

 violently

0.07

 villa

0.06

(ord

0.06

 puppy

0.06

_download

0.06

 dumpster

0.06

 coordinator

0.06

 powers

0.06

闪

0.06

 Bowl

0.06

Activations Density 0.000%

EST

This neuron primarily lights up on common short function words—prepositions and articles—rather than on substantive content terms.

references to statistical data and probabilities related to frequency and averages.

No Comments

No Known Activations

EST

This neuron primarily lights up on common short function words—prepositions and articles—rather than on substantive content terms.

references to statistical data and probabilities related to frequency and averages.

No Comments

No Known Activations