INDEX

Explanations

references to qualities and classifications

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

sidemargin

-0.64

bootstrapcdn

-0.61

 архивлан

-0.50

__(/*!

-0.48

 חיצוניים

-0.48

#+#

-0.47



-0.47

⤹

-0.46

 EdgeInsets

-0.46

Diweddarwch

-0.46

POSITIVE LOGITS

 kyse

0.46

velse

0.42

 hipó

0.35

طلحات

0.35

 aikana

0.34

 TextAppearance

0.32

 fotográfico

0.32

zeczytaj

0.31

isasi

0.30

Principles

0.30

Activations Density 0.677%

references to qualities and classifications

No Comments

No Known Activations