INDEX

Explanations

terms related to wearing and the experience of the wearer

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

featureID

-0.47



-0.43

 têtes

-0.42

ⓧ

-0.41

mybatisplus

-0.40

CreateModel

-0.39

ConverterFactory

-0.39

 medži

-0.38

IntoConstraints

-0.37

 materials

-0.37

POSITIVE LOGITS

CloseOperation

0.56

sidemargin

0.52

SequentialGroup

0.45

ANTLR

0.44

StartTag

0.42

naked

0.42

slidesToShow

0.42

gonic

0.42

DataSetChanged

0.41

runApp

0.40

Activations Density 0.005%

terms related to wearing and the experience of the wearer

No Comments

No Known Activations