INDEX

Explanations

computer code

np_max-act · gemini-2.0-flash

This neuron chiefly activates on numeric literals (including integers and floating-point constants) in code.

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

THOOK

-0.07

�

-0.06

 makeshift

-0.06

.DataVisualization

-0.06

gag

-0.06

.easing

-0.06

 BitConverter

-0.06

	left

-0.06

 simplest

-0.06

 слиз

-0.06

POSITIVE LOGITS

 trách

0.08

раст

0.07

 SORT

0.07

OME

0.07

 khai

0.06

аша

0.06

 renovations

0.06

品

0.06

ालन

0.06

SEM

0.06

Activations Density 0.006%

computer code

This neuron chiefly activates on numeric literals (including integers and floating-point constants) in code.

No Comments

No Known Activations