INDEX

Explanations

references to power and its various characteristics or 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 @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

<![

-0.47

jface

-0.44

 betweenstory

-0.44

ECM

-0.42

ViewInit

-0.42

 disagreeable

-0.41

 annals

-0.41

withIdentifier

-0.41

 juxta

-0.39

ftagPool

-0.39

POSITIVE LOGITS

 power

2.34

 Power

2.28

Power

2.25

power

2.19

 POWER

2.05

POWER

1.99

 powers

1.62

 Powers

1.59

Powers

1.57

powers

1.56

Activations Density 0.134%

references to power and its various characteristics or contexts

No Comments

No Known Activations