INDEX

Explanations

terms and phrases related to views, filters, and hierarchical structures within data processing contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 CreateTagHelper

-0.40

]-->

-0.36

 AdapterView

-0.34

MBC

-0.32

makeText

-0.31

MBC

-0.31

 brick

-0.31

chartInstance

-0.31

IntoConstraints

-0.31

uss

-0.30

POSITIVE LOGITS

IVEREF

0.60

цездатний

0.54

StoreMessageInfo

0.54

 okuyayım

0.51

Carriera

0.50

 noDo

0.50

ativement

0.48

sizeCache

0.48

verwijspagina

0.47

oa̍t

0.47

Activations Density 0.284%

terms and phrases related to views, filters, and hierarchical structures within data processing contexts

No Comments

No Known Activations