INDEX

Explanations

references to guidance or assistance related to tasks or subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

ốt

-0.56

principalTable

-0.49

rufen

-0.47

 disease

-0.42

Veuillez

-0.40

Crossref

-0.40

essoas

-0.40

 Архів

-0.38

Kedua

-0.38

olm

-0.37

POSITIVE LOGITS

Personendaten

0.83

contentLoaded

0.75

новништво

0.70

 chré

0.68

 ویکی‌پدیا

0.65

MenuGroup

0.63

évaluateur

0.63

 Chwiliwch

0.62

 ویکی‌پدی

0.62

featureID

0.61

Activations Density 0.360%

references to guidance or assistance related to tasks or subjects

No Comments

No Known Activations