INDEX

Explanations

phrases indicating appropriateness or correctness in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

createNewFile

-0.45

 commonly

-0.44

apatalk

-0.39

-|-

-0.38

wiec

-0.38

★☆

-0.38

Cyfarwyddwr

-0.38

Vidite

-0.37

digd

-0.37

Danh

-0.37

POSITIVE LOGITS

 wrong

1.01

 Wrong

0.98

Wrong

0.95

 giusta

0.94

wrong

0.93

 Correct

0.90

 richtigen

0.89

 correct

0.89

 WRONG

0.88

Correct

0.85

Activations Density 0.268%

phrases indicating appropriateness or correctness in various contexts

No Comments

No Known Activations

phrases indicating appropriateness or correctness in various contexts

No Comments

No Known Activations