INDEX

Explanations

Finding and stopping

np_max-act · gemini-2.0-flash

The neuron fires on verbs that signal resolving a conflict or completing an objective—words like “solve,” “find,” “neutralize,” and “save.”

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

 ifad

-0.07

asure

-0.07

	node

-0.07

(log

-0.06

 interact

-0.06

 Strait

-0.06

 Bran

-0.06

Nep

-0.06

variant

-0.06

wes

-0.06

POSITIVE LOGITS

 BrowserRouter

0.07

 coquine

0.06

:UIControlStateNormal

0.06

AuthProvider

0.06

.PLAIN

0.06

layarak

0.06

.WARNING

0.06

ाइल

0.06

award

0.06

 commodo

0.06

Activations Density 0.045%

Finding and stopping

The neuron fires on verbs that signal resolving a conflict or completing an objective—words like “solve,” “find,” “neutralize,” and “save.”

No Comments

No Known Activations