INDEX

Explanations

references to significant geographical, historical, or organizational facts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ujednoznacz

-0.88

 فريبيس

-0.81

躇

-0.78

 تضيفلها

-0.73

 متعلقه

-0.70

 незавершена

-0.69

 الرياضيه

-0.68

joueurs

-0.68

endenza

-0.67

inflater

-0.67

POSITIVE LOGITS

也是

0.61

 ever

0.52

it

0.50

 also

0.49

 EVER

0.48

 überhaupt

0.47

 sauvages

0.45

principalColumn

0.44

꼽

0.44

due

0.43

Activations Density 0.247%

references to significant geographical, historical, or organizational facts

No Comments

No Known Activations