INDEX

Explanations

numerical data and references to academic publications

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_16/width_16k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

]")]

-0.59

 Paglinawan

-0.59

 snippetHide

-0.56

crollView

-0.56

 iprot

-0.49

ſelf

-0.47

 lagerfeld

-0.47

]='\

-0.47

 }}"></

-0.46

 autorytatywna

-0.44

POSITIVE LOGITS

Подробнее

0.39

ev

0.36

 Submitted

0.35

 pending

0.35

 submitted

0.35

 submission

0.35

unmodifiable

0.34

ges

0.34

Submitted

0.34

 sync

0.34

Activations Density 0.803%

numerical data and references to academic publications

No Known Activations