INDEX

Explanations

phrases indicating suitability or recommendations

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

 dziel

-0.30

 répar

-0.29

 przez

-0.29

 compañ

-0.27

 instrucciones

-0.26

</h6>

-0.26

 consejo

-0.25

 auprès

-0.25

설

-0.25

 huellas

-0.25

POSITIVE LOGITS

NameInMap

0.76

niſſe

0.73

 Verſ

0.70

 laſſen

0.68

 queſto

0.68

 ſich

0.68

 unſer

0.68

iſen

0.67

 Pardavimas

0.67

 ſind

0.66

Activations Density 0.124%

phrases indicating suitability or recommendations

No Comments

No Known Activations