INDEX

Explanations

reasons, explanations

np_max-act · gemini-2.0-flash

The neuron activates on words that signal causal or explanatory relations (e.g. “reason,” “source,” “explain”).

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

 evaluating

-0.07

-Am

-0.07

oler

-0.07

 envelope

-0.06

_symbol

-0.06

oo

-0.06

улю

-0.06

 inhibit

-0.06

ec

-0.06

_LEFT

-0.06

POSITIVE LOGITS

.getDrawable

0.07

ระยะ

0.06

ensagem

0.06

سبة

0.06

 грн

0.06

 Args

0.06

rdf

0.06

ARCHAR

0.06

byl

0.06

=:

0.06

Activations Density 0.049%

reasons, explanations

The neuron activates on words that signal causal or explanatory relations (e.g. “reason,” “source,” “explain”).

No Comments

No Known Activations