INDEX

Explanations

)

np_max-act · gemini-2.0-flash

The neuron detects metadata lines, especially numeric sequences and date‐ or time‐style tokens (e.g. years, timestamps, version numbers) at the beginnings of text segments.

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

 noticing

-0.07

 posing

-0.07

 playful

-0.07

 Furious

-0.07

 verschied

-0.06

 winter

-0.06

WillAppear

-0.06

Ω

-0.06

 rivers

-0.06

Being

-0.06

POSITIVE LOGITS

obutton

0.06

 globe

0.06

 incred

0.06

NSE

0.06

flatMap

0.06

她的

0.06

중

0.06

 нем

0.06

(let

0.05

Activations Density 0.040%

)

The neuron detects metadata lines, especially numeric sequences and date‐ or time‐style tokens (e.g. years, timestamps, version numbers) at the beginnings of text segments.

No Comments

No Known Activations