INDEX

Explanations

always

np_max-act · gemini-2.0-flash

The neuron activates on tokens that give prescriptive safety or compliance advice—especially “always,” “follow,” and similar guideline-style recommendations.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

repr

-0.07

 heartfelt

-0.07

Smy

-0.07

 پار

-0.06

mx

-0.06

드립니다

-0.06

 trou

-0.06

GetWidth

-0.06

Syn

-0.06

FFT

-0.06

POSITIVE LOGITS

 deploy

0.07

 μη

0.07

 Arabia

0.07

 communications

0.06

 eros

0.06

 Chain

0.06

 minimise

0.06

 combine

0.06

_init

0.06

 Consulting

0.06

Activations Density 0.017%

always

The neuron activates on tokens that give prescriptive safety or compliance advice—especially “always,” “follow,” and similar guideline-style recommendations.

No Comments

No Known Activations