INDEX

Explanations

references to historical contexts and figures associated with ancient Roman culture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ſelves

-0.75

ſelf

-0.73

 myſelf

-0.72

 Jefus

-0.71

 unſ

-0.69

 purpoſe

-0.69

 Majefty

-0.69

Eſ

-0.68

 Diſ

-0.68

 Reſ

-0.67

POSITIVE LOGITS

 romanos

0.51

 civilización

0.44

 römischen

0.43

 desnuda

0.42

 Danach

0.41

ladiator

0.41

 griego

0.40

 mengenal

0.39

 pasaba

0.37

detalle

0.37

Activations Density 0.251%

references to historical contexts and figures associated with ancient Roman culture

No Comments

No Known Activations