INDEX

Explanations

Thinking

np_max-act · gemini-2.0-flash

This neuron lights up on words and phrases that express internal deliberation or uncertainty—e.g. “think,” “consider,” “know,” “decide,” and similar language of weighing options.

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

 biology

-0.08

 triggering

-0.06

_KEYWORD

-0.06

ılması

-0.06

_MEMBERS

-0.06

�

-0.06

 모습

-0.06

 rencont

-0.06

บอก

-0.06

Ohio

-0.06

POSITIVE LOGITS

ература

0.08

είται

0.07

-lib

0.07

(actor

0.06

-Bar

0.06

 consequence

0.06

 Lists

0.06

Reject

0.06

 Students

0.06

Activations Density 0.026%

Thinking

This neuron lights up on words and phrases that express internal deliberation or uncertainty—e.g. “think,” “consider,” “know,” “decide,” and similar language of weighing options.

No Comments

No Known Activations

Thinking

This neuron lights up on words and phrases that express internal deliberation or uncertainty—e.g. “think,” “consider,” “know,” “decide,” and similar language of weighing options.

No Comments

No Known Activations