INDEX

Explanations

the

np_max-act · gemini-2.0-flash

The neuron spotlights mentions of historical wars and conflicts—activating on war names and their associated dates in battle descriptions.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

references to specific historical battles.

oai_token-act-pair · gpt-4o-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

عت

-0.07

du

-0.06

 firepower

-0.06

 Medina

-0.06

 над

-0.06

 delegation

-0.06

eax

-0.06

(mx

-0.06

 труд

-0.06

_W

-0.05

POSITIVE LOGITS

 repmat

0.07

 viene

0.07

 transporter

0.07

_accessible

0.06

 Plains

0.06

getDefault

0.06

 Contributor

0.06

ísk

0.06

-formed

0.06

 которых

0.06

Activations Density 0.021%

the

The neuron spotlights mentions of historical wars and conflicts—activating on war names and their associated dates in battle descriptions.

references to specific historical battles.

No Comments

No Known Activations