INDEX

Explanations

tags and structural components in a document, particularly HTML and XML-like markup

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ruptedException

-0.70

 Inſ

-0.64

 Anſ

-0.62

 Reſ

-0.62

 themſelves

-0.60

 queſta

-0.58

wapV

-0.57

 habet

-0.56

MLLoader

-0.56

 Diſ

-0.56

POSITIVE LOGITS

 autorytatywna

0.46

 acronym

0.46

 siglas

0.42

RTDA

0.42

 Italijani

0.35

ShortName

0.33

Odkazy

0.33

 abbreviation

0.33

 acronyms

0.32

Thx

0.31

Activations Density 2.613%

tags and structural components in a document, particularly HTML and XML-like markup

No Comments

No Known Activations