INDEX

Explanations

Aimed at specific audience

np_max-act · gemini-2.0-flash

The neuron responds to bits of instructional or directive text—it picks up on language framing explanations or commands (e.g. “meant for,” “dirigidas a,” “direcionadas aos”), signaling procedural or teaching‐style sentences.

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

۱۹۴

-0.06

र

-0.06

ون

-0.06

 Coloring

-0.06

Ng

-0.06

 قب

-0.06

PUR

-0.06

 нач

-0.06

_pet

-0.06

POSITIVE LOGITS

�

0.07

惠

0.07

compute

0.07

iParam

0.07

rections

0.07

--;

0.06

 Nations

0.06

unteer

0.06

.integer

0.06

Calibri

0.06

Activations Density 0.182%

Aimed at specific audience

The neuron responds to bits of instructional or directive text—​it picks up on language framing explanations or commands (e.g. “meant for,” “dirigidas a,” “direcionadas aos”), signaling procedural or teaching‐style sentences.

No Comments

No Known Activations

The neuron responds to bits of instructional or directive text—it picks up on language framing explanations or commands (e.g. “meant for,” “dirigidas a,” “direcionadas aos”), signaling procedural or teaching‐style sentences.