INDEX

Explanations

terms related to restrictions or prohibitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

sidemargin

-0.79

 Chwiliwch

-0.79

MLLoader

-0.77

ThroughAttribute

-0.68

 GenerationType

-0.67

évaluateur

-0.66

र्भ

-0.64

-0.63

PhysRev

-0.63

 addSubview

-0.63

POSITIVE LOGITS

 prohibited

2.80

 forbidden

2.75

 banned

2.67

 prohibition

2.42

 forbid

2.36

 forbade

2.25

forbidden

2.25

 forbids

2.23

 prohibit

2.22

 prohibits

2.22

Activations Density 0.047%

terms related to restrictions or prohibitions

No Comments

No Known Activations