INDEX

Explanations

variations of the word "composition."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 inoxidable

-0.39

 boletín

-0.39

 embuti

-0.38

 telefónica

-0.38

 continuidad

-0.38

ワイイ

-0.38

 prédé

-0.37

 gustado

-0.36

+#+

-0.36

 vettore

-0.36

POSITIVE LOGITS

yo

0.80

rungsseite

0.79

Yo

0.77

yo

0.77

Composition

0.75

 Composition

0.75

Yo

0.71

 composition

0.70

yi

0.69

composition

0.69

Activations Density 0.271%

variations of the word "composition."

No Comments

No Known Activations