INDEX

Explanations

occurrences of the word "die" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 TextAppearance

-0.88

 Jurí

-0.65

 Tenga

-0.64

 Meck

-0.63

<!--[

-0.60

 clemen

-0.59

 borrower

-0.59

 censi

-0.58

 averted

-0.57

 hierogly

-0.57

POSITIVE LOGITS

Die

1.27

die

1.26

Die

1.17

0.95

DIE

0.94

die

0.88

DIE

0.77

Unsere

0.75

 Diese

0.71

Kanpo

0.68

Activations Density 0.042%

occurrences of the word "die" in various contexts

No Comments

No Known Activations