INDEX

Explanations

code/compiler outputs

np_max-act · gemini-2.0-flash

The neuron fires strongly on assembler directive tokens (those beginning with a “.” such as .thumb, .type, .alias, .syntax, etc.) and related metadata labels.

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

.getB

-0.07

 FIFO

-0.07

 engage

-0.06

假

-0.06

 Goodman

-0.06

 каф

-0.06

"F

-0.06

 Barry

-0.06

Ide

-0.06

.getS

-0.06

POSITIVE LOGITS

说明

0.06

่าง

0.06

//_

0.06

 sapi

0.06

 hollow

0.06

jes

0.06

_paper

0.06

↵↵

0.06

rchive

0.06

 recognised

0.06

Activations Density 0.005%

code/compiler outputs

The neuron fires strongly on assembler directive tokens (those beginning with a “.” such as .thumb, .type, .alias, .syntax, etc.) and related metadata labels.

No Known Activations

code/compiler outputs

The neuron fires strongly on assembler directive tokens (those beginning with a “.” such as .thumb, .type, .alias, .syntax, etc.) and related metadata labels.

No Known Activations