INDEX

Explanations

references to methodology sections in research or academic papers

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

getDoctrine

-0.51

 autorytatywna

-0.51

AsUp

-0.50

Aiheesta

-0.50

 saites

-0.47

Зноскі

-0.44

 suicidio

-0.42

—

-0.42

hésite

-0.42

 wuß

-0.42

POSITIVE LOGITS

AndEndTag

0.79

InjectAttribute

0.69

ResumeLayout

0.69

ValueStyle

0.65

GEBURTSDATUM

0.64

Hozzáférés

0.57

 viewDidLoad

0.57

منابع

0.56

وردار

0.55

Activations Density 0.000%

references to methodology sections in research or academic papers

No Known Activations