INDEX

Explanations

code elements related to user management and session handling

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

ProductCategory

-0.56

出版年

-0.52

OrderService

-0.50

gericht

-0.47

ValueStyle

-0.46

CategoryId

-0.45

 ModelExpression

-0.44

ConstraintMaker

-0.42

 оригіналу

-0.42

IVEREF

-0.41

POSITIVE LOGITS

 person

1.39

person

1.20

 user

1.19

 persons

1.19

 Person

1.18

Person

1.14

 users

1.14

 PERSON

1.12

 personne

1.11

 persona

1.11

Activations Density 0.307%

code elements related to user management and session handling

No Comments

No Known Activations