INDEX

Explanations

"

np_max-act · gemini-2.0-flash

This neuron fires primarily on numeric tokens (e.g., dates, times, week numbers).

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

raf

-0.06

 Engel

-0.06

+k

-0.06

 wipe

-0.06

 Marriage

-0.06

.ReadToEnd

-0.06

genden

-0.06

ekkür

-0.06

계

-0.06

特

-0.06

POSITIVE LOGITS

 на

0.07

 flick

0.07

 Glass

0.06

"):

0.06

 collegiate

0.06

 specialists

0.06

.setStroke

0.06

 sembl

0.06

.addComponent

0.06

 colonization

0.06

Activations Density 0.277%

"

This neuron fires primarily on numeric tokens (e.g., dates, times, week numbers).

No Comments

No Known Activations

"

This neuron fires primarily on numeric tokens (e.g., dates, times, week numbers).

No Comments

No Known Activations