INDEX

Explanations

structured data representations or markup syntax within a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

WithMany

-0.72

 ویکی‌پدیا

-0.49

 einz

-0.36

andReturn

-0.36

amoan

-0.35

 الرخصة

-0.34

kuuta

-0.34

แห่ง

-0.33

Pozri

-0.33

 Usaha

-0.33

POSITIVE LOGITS

expandindo

0.73

 ब्रेकडाउन

0.59

IMC

0.57

 defaultstate

0.56

TokenNameLBRACE

0.54

IUrlHelper

0.52

bmi

0.49

rbrakk

0.49

səhifə

0.49

TagHelper

0.48

Activations Density 2.441%

structured data representations or markup syntax within a document

No Comments

No Known Activations