INDEX

Explanations

legal terminology and references to evidence and defendants in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jih

-0.51

otechnical

-0.49

ReusableCell

-0.47

 gouvern

-0.46

zygous

-0.45

 призна

-0.45

dera

-0.45

くると

-0.44

ddling

-0.44

dd

-0.44

POSITIVE LOGITS

ArgsConstructor

0.79

 StatelessWidget

0.75

NUMX

0.74

saraba

0.74

SequentialGroup

0.72

]")]

0.71

 Chwiliwch

0.69

IsMutable

0.67

 autorytatywna

0.66

Personensuche

0.66

Activations Density 0.291%

legal terminology and references to evidence and defendants in legal contexts

No Comments

No Known Activations