INDEX

Explanations

indicators of performance and progress in competitive scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

:✨

-0.57

 avoient

-0.57

 littéraire

-0.52

 المعيارى

-0.50

 disambiguazione

-0.49

 éduc

-0.48

 étoient

-0.47

AddressBook

-0.47

 مشين

-0.46

écriture

-0.46

POSITIVE LOGITS

ChildScrollView

0.47

ISupport

0.45

icitis

0.40

 <<<<<<<<<<<<<<

0.38

requipa

0.35

 opérateur

0.35

 negoti

0.34

 Intere

0.33

 remonter

0.33

WithIOException

0.33

Activations Density 0.005%

indicators of performance and progress in competitive scenarios

No Comments

No Known Activations