INDEX

Explanations

terms related to compositions and mixtures in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 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

Jeografia

-0.70

 ainfi

-0.69

 חיצוניים

-0.66

PhysRevLett

-0.66

 feroit

-0.66

 utafitiHapana

-0.65

 plufieurs

-0.64

WireFormatLite

-0.62

 auroit

-0.61

 étoient

-0.60

POSITIVE LOGITS

 composition

1.18

 composite

1.11

 compose

1.08

 composed

1.05

Compose

1.02

 Compose

1.01

 Composition

1.00

 composing

0.99

 COMPOS

0.96

 compositions

0.95

Activations Density 0.247%

terms related to compositions and mixtures in various contexts

No Comments

No Known Activations