INDEX

Explanations

specific categories of nouns and topics related to medical issues and entertainment

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 autorytatywna

-0.59

ariConfig

-0.54

 informée

-0.51

úgó

-0.51

ModelAdmin

-0.51

IActionResult

-0.50

tomans

-0.50

handleRequest

-0.50

 مشين

-0.50

eleste

-0.50

POSITIVE LOGITS

 ویکی‌پدی

0.46

Personensuche

0.41

PyExc

0.35

msgTypes

0.31

etc

0.30

ChildScrollView

0.30

hyrchwyd

0.29

etc

0.29

 مرئيه

0.28

 newArray

0.28

Activations Density 0.031%

specific categories of nouns and topics related to medical issues and entertainment

No Comments

No Known Activations