INDEX

Explanations

references to chapters, fictional content, and personal messages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

RegressionTest

-0.53

Portail

-0.53

 ModelExpression

-0.51

TestingModule

-0.47

ToBounds

-0.46

客様

-0.45

たしか

-0.43

ContentAlignment

-0.42

Zeneca

-0.42

原始内容存档于

-0.42

POSITIVE LOGITS

 hope

0.81

 Enjoy

0.75

 ENJOY

0.73

Hope

0.72

 enjoy

0.72

 Hope

0.71

 disclaimer

0.71

 hoped

0.70

 hopefully

0.69

enjoy

0.67

Activations Density 0.413%

references to chapters, fictional content, and personal messages

No Comments

No Known Activations