INDEX

Explanations

phrases related to legal defenses, particularly in the context of self-defense and mitigating circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

ppins

-0.51

 contempt

-0.49

ingu

-0.45

Населення

-0.45

epam

-0.45

ugou

-0.45

 contemp

-0.43

انظر

-0.43

 gespannt

-0.42

fseek

-0.42

POSITIVE LOGITS

 unintentional

0.79

 jspb

0.75

 excus

0.75

 InputDecoration

0.69

LEncoder

0.69

 مشين

0.69

RectangleBorder

0.68

 propOrder

0.67

 innocent

0.67

UnsafeEnabled

0.66

Activations Density 0.462%

phrases related to legal defenses, particularly in the context of self-defense and mitigating circumstances

No Comments

No Known Activations