INDEX

Explanations

terms related to anti-activities or anti-drugs

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

against

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

期刊论文

-0.65

SBATCH

-0.54

cased

-0.53

impanan

-0.52

ellyn

-0.49

躇

-0.49

owning

-0.47

ableView

-0.47

 RoutedEventArgs

-0.47

featureID

-0.47

POSITIVE LOGITS

 anti

1.18

 Anti

1.08

Anti

0.94

 ANTI

0.93

 Анти

0.82

anti

0.79

 анти

0.78

Анти

0.72

 antim

0.71

ANTI

0.68

Activations Density 0.029%

terms related to anti-activities or anti-drugs

against

No Comments

No Known Activations