INDEX

Explanations

references to changes, modifications, and replacements in code or configuration 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-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

AutoScaleMode

-0.52

ArgsConstructor

-0.49

AsUp

-0.44

المناصب

-0.43

featureID

-0.42

Descrizione

-0.40

colnshire

-0.39

(~(

-0.39

はじめに

-0.38

 funniest

-0.38

POSITIVE LOGITS

 SwitchCompat

0.45

didSet

0.40

Història

0.40

 cantit

0.40

IContainer

0.40

 ***!

0.39

 rechange

0.38

 perioada

0.37

HÀ

0.36

หมาย

0.35

Activations Density 0.568%

references to changes, modifications, and replacements in code or configuration contexts

No Comments

No Known Activations