INDEX

Explanations

quotations

np_max-act · gemini-2.0-flash

The neuron strongly activates on polite, self-referential disclaimer phrases—especially “I would like to remind…” style first-person statements.

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

 residency

-0.07

 násled

-0.07

ETIME

-0.06

 resultMap

-0.06

 홈페이지

-0.06

 سنت

-0.06

.)↵↵↵↵

-0.06

 Tamil

-0.06

 fakat

-0.06

 rele

-0.06

POSITIVE LOGITS

ulous

0.07

 elektrik

0.07

 Wouldn

0.07

 tailor

0.07

 dishwasher

0.07

回

0.06

 Establish

0.06

 Cliff

0.06

 nghề

0.06

 stylish

0.06

Activations Density 0.003%

quotations

The neuron strongly activates on polite, self-referential disclaimer phrases—especially “I would like to remind…” style first-person statements.

No Comments

No Known Activations

quotations

The neuron strongly activates on polite, self-referential disclaimer phrases—especially “I would like to remind…” style first-person statements.

No Comments

No Known Activations