INDEX

Explanations

references to recent developments, studies, or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

oredCriteria

-0.67

 pleaſure

-0.60

 myſelf

-0.53

ValueGenerated

-0.52

 faſt

-0.52

 ddelweddau

-0.49

 purpoſe

-0.49

AxisAlignment

-0.47

 ſta

-0.46

GEBURTSDATUM

-0.45

POSITIVE LOGITS

 recent

2.56

recent

2.47

Recent

2.38

 Recent

2.28

 RECENT

2.14

 recently

2.00

RECENT

2.00

 reciente

1.98

recently

1.96

 recente

1.92

Activations Density 0.101%

references to recent developments, studies, or reports

No Comments

No Known Activations