INDEX

Explanations

a

np_max-act · gemini-2.0-flash

discussions surrounding updates and changes in software or organizational policies.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

This neuron lights up on evaluative or emphatic words—adjectives and adverbs that convey an opinion or emphasis (e.g. “good,” “true,” “surely,” “especially,” “largely”).

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

ilig

-0.07

}:

-0.07

amber

-0.06

 Belarus

-0.06

_customize

-0.06

 forCellReuseIdentifier

-0.06

 Story

-0.06

 Adobe

-0.06

 faith

-0.06

 finale

-0.06

POSITIVE LOGITS

�

0.06

lendirme

0.06

,Th

0.06

_UNUSED

0.06

Pi

0.06

ッ

0.06

アメリカ

0.06

blocking

0.06

.contentType

0.06

PI

0.06

Activations Density 0.140%

a

discussions surrounding updates and changes in software or organizational policies.

This neuron lights up on evaluative or emphatic words—adjectives and adverbs that convey an opinion or emphasis (e.g. “good,” “true,” “surely,” “especially,” “largely”).

No Comments

No Known Activations

a

discussions surrounding updates and changes in software or organizational policies.

This neuron lights up on evaluative or emphatic words—adjectives and adverbs that convey an opinion or emphasis (e.g. “good,” “true,” “surely,” “especially,” “largely”).

No Comments

No Known Activations