INDEX

Explanations

arguments and philosophy

np_max-act · gemini-2.0-flash

The neuron activates on the language of formal philosophical argumentation—especially tokens related to causal or “first-mover” reasoning and structured premise-conclusion discourse.

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

 EXPER

-0.06

olynomial

-0.06

ENCHMARK

-0.06

 dévelop

-0.06

.Large

-0.06

ition

-0.06

 soils

-0.06

论

-0.06

 VARIABLE

-0.06

kn

-0.06

POSITIVE LOGITS

 Ultra

0.08

ồn

0.07

 وان

0.06

asını

0.06

uther

0.06

 Toshiba

0.06

排

0.06

apis

0.06

тах

0.06

CppTypeDefinition

0.06

Activations Density 0.029%

arguments and philosophy

The neuron activates on the language of formal philosophical argumentation—especially tokens related to causal or “first-mover” reasoning and structured premise-conclusion discourse.

No Comments

No Known Activations