INDEX

Explanations

keywords related to a specific programming component or module named "mantis."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

IPAC

-0.82

GEBURTSDATUM

-0.79

>{@

-0.78

PerformLayout

-0.75

erializer

-0.74

ymce

-0.73

بوابة

-0.71

principalTable

-0.69

mobileqq

-0.68

 ✭✭

-0.67

POSITIVE LOGITS

 itſelf

1.46

 myſelf

1.41

 Majefty

1.39

 himſelf

1.38

 themſelves

1.34

 whoſe

1.33

 Jefus

1.33

 ſtate

1.29

 Reſ

1.26

 becauſe

1.25

Activations Density 4.362%

keywords related to a specific programming component or module named "mantis."

No Comments

No Known Activations