INDEX

Explanations

Plus

np_max-act · gemini-2.0-flash

The neuron spotlights multiword proper names or titles (i.e. sequences of capitalized tokens forming named entities).

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

	save

-0.08

 mối

-0.07

 Schwe

-0.07

 tanım

-0.07

 save

-0.07

-key

-0.07

 studi

-0.07

save

-0.06

/<?

-0.06

 Detected

-0.06

POSITIVE LOGITS

 sublist

0.06

rcode

0.06

 odom

0.06

یم

0.06

lights

0.06

KeyUp

0.05

ecera

0.05

oking

0.05

속

0.05

 stopping

0.05

Activations Density 0.535%

Plus

The neuron spotlights multiword proper names or titles (i.e. sequences of capitalized tokens forming named entities).

No Comments

No Known Activations