INDEX

Explanations

the term "adjacent" and its variations, indicating relationships or proximity

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 MenuView

-0.45

MemoryWarning

-0.44

Pint

-0.44

MLLoader

-0.42

庁

-0.40

 ControllerBase

-0.40

 informée

-0.39

msgTypes

-0.39

pint

-0.38

caref

-0.38

POSITIVE LOGITS

 adjacent

2.31

 Adjacent

2.06

 neighboring

1.97

 neighbouring

1.93

adjacent

1.91

Adjacent

1.89

 nearby

1.81

 neighbor

1.77

 adjoining

1.77

neighbor

1.76

Activations Density 0.201%

the term "adjacent" and its variations, indicating relationships or proximity

No Comments

No Known Activations

the term "adjacent" and its variations, indicating relationships or proximity

No Comments

No Known Activations