INDEX

Explanations

contextual information about locations and entities

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

 bileklik

-0.94

 küpe

-0.93

 kılıf

-0.93

 kağıdı

-0.88

 taşlı

-0.85

 eşofman

-0.83

 hırka

-0.82

 kağı

-0.81

 desenli

-0.79

 kolye

-0.77

POSITIVE LOGITS

 متعلقه

0.92

 Turkey

0.77

 Turkish

0.75

 Istanbul

0.71

Turkey

0.71

İ

0.69

İ

0.68

 resourceCulture

0.66

Turkish

0.65

Gü

0.64

Activations Density 0.147%

contextual information about locations and entities

No Comments

No Known Activations