INDEX

Explanations

instances of structured data representation or formatting, particularly in programming or technical documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

<eos>

-0.92

featureID

-0.90

TestingModule

-0.76

astéro

-0.75

sizeCache

-0.70

 архивлан

-0.69

GOTREF

-0.69

كويكب

-0.68

okovic

-0.66

 textStatus

-0.65

POSITIVE LOGITS

urlpatterns

0.50

 Italijanski

0.49

jores

0.49

djangoproject

0.47

mybatisplus

0.44

twimg

0.42

↵↵

0.41

 Komisi

0.41

Publié

0.41

ANDOM

0.41

Activations Density 0.840%

instances of structured data representation or formatting, particularly in programming or technical documentation

No Comments

No Known Activations