INDEX

Explanations

classification

np_max-act · gemini-2.0-flash

The neuron consistently lights up on technical terms and stems related to distinguishing, classifying, or determining categories—e.g. “discriminate,” “denomination,” or “classify.”

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

Birthday

-0.06

fdb

-0.06

-conf

-0.06

 ENUM

-0.06

Miss

-0.06

 setMessage

-0.06

/locale

-0.06

.number

-0.06

 Directive

-0.06

 ridiculous

-0.06

POSITIVE LOGITS

 temporarily

0.07

nění

0.07

 relie

0.07

atient

0.06

.engine

0.06

 зовсім

0.06

bagai

0.06

ublic

0.06

载

0.06

ke

0.06

Activations Density 0.034%

classification

The neuron consistently lights up on technical terms and stems related to distinguishing, classifying, or determining categories—e.g. “discrim­inate,” “denomination,” or “classify.”

No Comments

No Known Activations

The neuron consistently lights up on technical terms and stems related to distinguishing, classifying, or determining categories—e.g. “discriminate,” “denomination,” or “classify.”