INDEX

Explanations

implementation details

np_max-act · gemini-2.0-flash

This neuron detects discussion of compiler implementation behavior, especially mentions of what compilers are allowed or allowed not to do (e.g. optimizations, storage allocation, the “as-if” rule).

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

thickness

-0.06

ให

-0.06

 drum

-0.06

>Description

-0.06

 розрах

-0.06

ประถม

-0.06

 hạt

-0.06

 plates

-0.06

.carousel

-0.06

DECREF

-0.06

POSITIVE LOGITS

 parcels

0.07

 groundwater

0.07

 inspirational

0.06

„N

0.06

(FILE

0.06

_Grid

0.06

_ASSOC

0.06

 convey

0.06

quad

0.06

 diagram

0.06

Activations Density 0.031%

implementation details

This neuron detects discussion of compiler implementation behavior, especially mentions of what compilers are allowed or allowed not to do (e.g. optimizations, storage allocation, the “as-if” rule).

No Comments

No Known Activations