INDEX

Explanations

C code

np_max-act · gemini-2.0-flash

This neuron activates on tokens belonging to low‐level hardware I/O operations (e.g. port/memory read and write functions like inb, outb, insb, outsb, readq, writeq).

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

Negative Logits

Wer

-0.07

_workers

-0.06

iph

-0.06

许多

-0.06

Bind

-0.06

찮

-0.06

urga

-0.06

 Laboratories

-0.06

Wer

-0.06

 되었

-0.06

POSITIVE LOGITS

 추천

0.07

_difference

0.06

 freedoms

0.06

>=

0.06

 quat

0.06

ob

0.06

 desc

0.06

_APPS

0.06

aj

0.06

.',

0.06

Activations Density 0.006%

C code

This neuron activates on tokens belonging to low‐level hardware I/O operations (e.g. port/memory read and write functions like inb, outb, insb, outsb, readq, writeq).

No Known Activations

C code

This neuron activates on tokens belonging to low‐level hardware I/O operations (e.g. port/memory read and write functions like inb, outb, insb, outsb, readq, writeq).

No Known Activations