INDEX

Explanations

possibility/hypothetical situations

np_max-act · gemini-2.0-flash

The neuron responds to modal and speculative language—words expressing possibility, uncertainty, or commentary (e.g. could, would, think, likely).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aldo

-0.08

IH

-0.07

inters

-0.07

.IntegerField

-0.07

Five

-0.07

ué

-0.07

 Devil

-0.06

ılım

-0.06

凡

-0.06

.Abs

-0.06

POSITIVE LOGITS

 streamline

0.06

 бли

0.06

 складу

0.06

 confidential

0.06

Gri

0.06

 depart

0.06

 prohibiting

0.06

 Pandora

0.06

(gameObject

0.06

.connect

0.06

Activations Density 0.112%

possibility/hypothetical situations

The neuron responds to modal and speculative language—words expressing possibility, uncertainty, or commentary (e.g. could, would, think, likely).

No Comments

No Known Activations