INDEX

Explanations

path

np_max-act · gemini-2.0-flash

The neuron consistently activates on occurrences of the identifier “path” (e.g. require('path') or path.xxx), i.e. references to the file‐system path module or path operations.

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

rente

-0.07

็นส

-0.07

 ให

-0.07

ارش

-0.07

่าต

-0.07

-danger

-0.06

 Proud

-0.06

ونی

-0.06

 NOAA

-0.06

_WAIT

-0.06

POSITIVE LOGITS

amm

0.07

 intrinsic

0.07

uibModal

0.06

path

0.06

 日本

0.06

้าง

0.06

 назна

0.06

 количе

0.06

objc

0.06

 plagued

0.06

Activations Density 0.001%

path

The neuron consistently activates on occurrences of the identifier “path” (e.g. require('path') or path.xxx), i.e. references to the file‐system path module or path operations.

No Comments

No Known Activations

path

The neuron consistently activates on occurrences of the identifier “path” (e.g. require('path') or path.xxx), i.e. references to the file‐system path module or path operations.

No Comments

No Known Activations