INDEX

Explanations

questions and inquiries directed towards the reader

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

SBATCH

-0.72

########.

-0.72

makeConstraints

-0.71

 snippetHide

-0.69

ftagPool

-0.67

 iprot

-0.65

enterOuterAlt

-0.65

Diweddarwch

-0.65

bootstrapcdn

-0.65

verwijspagina

-0.64

POSITIVE LOGITS

0.35

 needs

0.33

 StatefulWidget

0.29

 domy

0.29

是不是

0.28

 doświad

0.28

or

0.28

你是不是

0.28

Quieres

0.27

Activations Density 0.014%

questions and inquiries directed towards the reader

No Comments

No Known Activations