INDEX

Explanations

instances of the word "add" and its variations, indicating a focus on inclusion or augmentation

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 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

maketitle

-0.82

findpost

-0.73

Bibliografia

-0.70

 LoggerFactory

-0.65

archiviato

-0.64

 Wikispecies

-0.63

 Bourgeois

-0.63

openConnection

-0.63

thansa

-0.63

 Италијани

-0.62

POSITIVE LOGITS

add

3.32

Add

3.04

 adds

2.93

Add

2.88

 adding

2.73

 Adds

2.71

add

2.65

ADD

2.61

 added

2.58

 Adding

2.56

Activations Density 0.094%

instances of the word "add" and its variations, indicating a focus on inclusion or augmentation

No Comments

No Known Activations