INDEX

Explanations

units of time

np_max-act · gemini-2.0-flash

The neuron flags numeric tokens—especially decimal-formatted values (e.g. “2.25”, “0.58”) and other numbers indicating quantities or durations.

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

少し

-0.07

监督

-0.07

veriş

-0.06

 DataView

-0.06

 Deals

-0.06

เหต

-0.06

.tsv

-0.06

 rendre

-0.06

 thinner

-0.06

esi

-0.06

POSITIVE LOGITS

 hours

0.08

Years

0.08

ago

0.08

 minutes

0.08

 months

0.08

 weeks

0.08

 seconds

0.08

 час

0.07

pte

0.07

wc

0.07

Activations Density 0.018%

units of time

The neuron flags numeric tokens—especially decimal-formatted values (e.g. “2.25”, “0.58”) and other numbers indicating quantities or durations.

No Comments

No Known Activations