INDEX

Explanations

universes/worlds

np_max-act · gemini-2.0-flash

The neuron activates on words that evoke multiple co-existing possibilities or branching realities—especially terms like “parallel,” “univers(es),” “possibilities,” and related outcome-oriented language.

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

.ShouldBe

-0.06

戴

-0.06

.addObject

-0.06

 große

-0.06

เด

-0.06

 discomfort

-0.06

織

-0.06

fi

-0.06

POSITIVE LOGITS

 getState

0.07

vfs

0.06

 decor

0.06

NodeType

0.06

 поперед

0.06

_READY

0.06

[word

0.06

#ab

0.06

 Quant

0.06

YYS

0.06

Activations Density 0.008%

universes/worlds

The neuron activates on words that evoke multiple co-existing possibilities or branching realities—especially terms like “parallel,” “univers(es),” “possibilities,” and related outcome-oriented language.

No Comments

No Known Activations

universes/worlds

The neuron activates on words that evoke multiple co-existing possibilities or branching realities—especially terms like “parallel,” “univers(es),” “possibilities,” and related outcome-oriented language.

No Comments

No Known Activations