INDEX

Explanations

special characters

np_max-act · gemini-2.0-flash

This neuron activates on raw assembly instruction tokens—mnemonics like mov/push/call and register or immediate syntax.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

婦

-0.07

bac

-0.07

็

-0.06

 FAIL

-0.06

READING

-0.06

 Ninth

-0.06

********************************************************

-0.06

 текущ

-0.06

_EDIT

-0.06

แนะ

-0.06

POSITIVE LOGITS

ubern

0.06

 طب

0.06

.Msg

0.06

 dehydration

0.06

#"

0.06

γε

0.06

 والتي

0.06

 onde

0.06

 statue

0.06

진

0.06

Activations Density 0.005%

special characters

This neuron activates on raw assembly instruction tokens—mnemonics like mov/push/call and register or immediate syntax.

No Comments

No Known Activations