INDEX

Explanations

terms related to detection methods and descriptor metrics in neural networks

oai_token-act-pair · gpt-4o-mini Triggered by @bot

detection and detectors

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 StatefulWidget

-0.49

hafter

-0.41

inspace

-0.41

+#+#

-0.41

setAccessible

-0.38

kre

-0.37

RegressionTest

-0.37

 onSave

-0.36

Jegyzetek

-0.36

󠁢

-0.35

POSITIVE LOGITS

ctron

0.47

apimachinery

0.47

Conventional

0.45

Vogue

0.45

 dran

0.43

 InputDecoration

0.43

ViewSet

0.43

Instagram

0.42

 hiér

0.41

unknownFields

0.41

Activations Density 0.012%

terms related to detection methods and descriptor metrics in neural networks

detection and detectors

No Comments

No Known Activations