INDEX

Explanations

references to achievements in publishing and notable accomplishments in various fields

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/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.27.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

ministrazione

-0.38

gezogen

-0.34

 elongated

-0.32

 Verständnis

-0.31

DAOImpl

-0.31

 règne

-0.31

delayed

-0.30

 fidèle

-0.30

 appassion

-0.30

 répertoire

-0.30

POSITIVE LOGITS

 prestigious

0.96

 famous

0.85

 celebrities

0.78

 renowned

0.78

 reputable

0.77

 ünl

0.75

famous

0.74

 major

0.74

 prominent

0.74

 elite

0.71

Activations Density 0.480%

references to achievements in publishing and notable accomplishments in various fields

No Comments

No Known Activations