INDEX

Explanations

references to sources or cited works in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

Personensuche

-1.01

 tartalomajánló

-1.00

 незавершена

-0.94

SharedDtor

-0.91

 Roskov

-0.86

IonicModule

-0.84

AsUp

-0.83

parsedMessage

-0.83

StoryboardSegue

-0.81

ьаж

-0.78

POSITIVE LOGITS

source

2.68

 source

1.98

Source

1.73

 Source

1.68

 SOURCE

1.56

SOURCE

1.53

ource

1.35

getSource

1.32

源

1.27

setSource

1.18

Activations Density 0.032%

references to sources or cited works in a text

No Comments

No Known Activations