INDEX

Explanations

legal terms and references related to court cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 виправивши

-1.02

IndentedString

-0.85

########.

-0.84

تقاوى

-0.79

 مرئيه

-0.77

 Roskov

-0.77

NameInMap

-0.76

featureID

-0.73

archiviato

-0.72

principalTable

-0.72

POSITIVE LOGITS

is

0.27

 屋根

0.26

好

0.26

 pres

0.25

ap

0.25

MD

0.25

 Editor

0.24

vo

0.24

 candidate

0.24

 time

0.24

Activations Density 0.830%

legal terms and references related to court cases

No Comments

No Known Activations