INDEX

Explanations

answering questions

np_max-act · gemini-2.0-flash

The main thing this neuron does is detect the assistant’s self-referential and hedging tokens when it introduces or qualifies information (e.g. “I can tell you that…”).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

words and phrases related to telecommunications technology and fiber optics.

oai_token-act-pair · gpt-4o-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

 ending

-0.06

 malformed

-0.06

-extension

-0.06

intros

-0.06

.UPDATE

-0.06

Bel

-0.06

 intimidating

-0.06

Kn

-0.06

 charitable

-0.06

 recovering

-0.06

POSITIVE LOGITS

 Boxes

0.07

_MT

0.07

 smarty

0.07

 trait

0.06

().'/

0.06

')),↵

0.06

ispiel

0.06

ाई

0.06

补

0.06

Activations Density 0.071%

answering questions

The main thing this neuron does is detect the assistant’s self-referential and hedging tokens when it introduces or qualifies information (e.g. “I can tell you that…”).

words and phrases related to telecommunications technology and fiber optics.

No Comments

No Known Activations