INDEX

Explanations

references to specific characters or elements within stories or lore

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

 respective

-0.39

 vriende

-0.38

 final

-0.37

final

-0.37

 initial

-0.36

 corresponding

-0.36

modelBuilder

-0.35

gesteld

-0.35

sieht

-0.34

Upgradeable

-0.34

POSITIVE LOGITS

 fallu

0.55

 виправивши

0.55

፩

0.51

 impunity

0.49

EndInit

0.48

ectoria

0.47

 betweenstory

0.46

 Photocase

0.46

 antics

0.45

Personensuche

0.45

Activations Density 0.942%

references to specific characters or elements within stories or lore

No Comments

No Known Activations