INDEX

Explanations

references to legal concepts and accountability in relation to actions and decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

 betweenstory

-0.51

CDCl

-0.42

 AssemblyCompany

-0.34

 lourdes

-0.34

tanleria

-0.33

 suiv

-0.33

Repeated

-0.33

 nonUne

-0.33

 oubli

-0.32

eradish

-0.32

POSITIVE LOGITS

 alive

0.87

 legible

0.76

 vulnerable

0.75

 liable

0.75

 compliant

0.73

 safe

0.73

 accessible

0.73

 accountable

0.73

 insolvent

0.71

 bankrupt

0.70

Activations Density 0.980%

references to legal concepts and accountability in relation to actions and decisions

No Comments

No Known Activations