INDEX

Explanations

descriptions and instructions

np_max-act · gemini-2.0-flash

The neuron detects words used to introduce explanatory or descriptive content—e.g. “help,” “explanation,” “outline,” “description,” etc.

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

DevExpress

-0.08

.JsonIgnore

-0.07

oper

-0.07

\a

-0.07

gard

-0.07

 breathe

-0.07

ableOpacity

-0.07

�

-0.06

 complication

-0.06

ić

-0.06

POSITIVE LOGITS

。',↵

0.06

赶

0.06

 Amen

0.06

 materia

0.06

@Service

0.06

 POST

0.06

，”

0.06

CELER

0.06

 ανά

0.06

RATE

0.06

Activations Density 0.022%

descriptions and instructions

The neuron detects words used to introduce explanatory or descriptive content—e.g. “help,” “explanation,” “outline,” “description,” etc.

No Comments

No Known Activations