INDEX

Explanations

quantifications and numerical data related to measurements and estimates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

 فريبيس

-1.03

 للمعارف

-0.80

:✨

-0.79

rungsseite

-0.79

PerformLayout

-0.75

rrggbb

-0.73

 Italijanski

-0.73

Демографія

-0.72

 HasFactory

-0.71

 arşivlendi

-0.69

POSITIVE LOGITS

 approximately

1.91

 roughly

1.67

approximately

1.57

 approx

1.48

 aproximadamente

1.42

roughly

1.41

 about

1.41

 Approximately

1.40

Approximately

1.33

 примерно

1.32

Activations Density 0.548%

quantifications and numerical data related to measurements and estimates

No Comments

No Known Activations