INDEX

Explanations

details about the quality and features of products, particularly in relation to their design, comfort, and aesthetic appeal

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

الحياه

-0.72

MemoryWarning

-0.72

ंदीखरीदारी

-0.71

تقاوى

-0.68

SequentialGroup

-0.68

Hentet

-0.67

</tfoot>

-0.66

 للاسماء

-0.64

 незавершена

-0.64

CWE

-0.64

POSITIVE LOGITS

 temperament

0.60

 exquis

0.57

 Exquisite

0.56

 delicate

0.51

 exquisite

0.50

 rate

0.47

 fresh

0.47

 charming

0.47

 Rate

0.47

 aura

0.47

Activations Density 0.055%

details about the quality and features of products, particularly in relation to their design, comfort, and aesthetic appeal

No Comments

No Known Activations