INDEX

Explanations

__

np_max-act · gemini-2.0-flash

The neuron flags C identifiers or keywords composed largely of built-in compiler macros (i.e. tokens containing double underscores like “__atomic_relaxed,” “attribute,” “__intptr_t,” etc.).

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

ricula

-0.07

 CHARSET

-0.07

cw

-0.06

sockopt

-0.06

 články

-0.06

 Summon

-0.06

Batch

-0.06

 worldly

-0.06

＿

-0.06

 icmp

-0.06

POSITIVE LOGITS

 sweat

0.06

respons

0.06

 whisper

0.06

ut

0.06

 исслед

0.06

 miss

0.06

.Product

0.06

 motivate

0.06

 Prop

0.06

+'_

0.06

Activations Density 0.004%

__

The neuron flags C identifiers or keywords composed largely of built-in compiler macros (i.e. tokens containing double underscores like “__atomic_relaxed,” “__attribute__,” “__intptr_t,” etc.).

No Comments

No Known Activations

The neuron flags C identifiers or keywords composed largely of built-in compiler macros (i.e. tokens containing double underscores like “__atomic_relaxed,” “attribute,” “__intptr_t,” etc.).