INDEX

Explanations

terms related to patent applications and legal proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 الحره

-0.51

 betweenstory

-0.48

lugin

-0.48

inigungs

-0.47

RuleContext

-0.47

StoreMessageInfo

-0.47

Благо

-0.46

beton

-0.46

 onData

-0.46

 Connell

-0.45

POSITIVE LOGITS

protoimpl

0.68

 invokingState

0.64

 Patent

0.59

 panto

0.58

Patent

0.56

 Walkover

0.56

 Invention

0.54

 Inventor

0.54

 INVENTION

0.54

asker

0.53

Activations Density 0.499%

terms related to patent applications and legal proceedings

No Comments

No Known Activations