INDEX

Explanations

terms and phrases related to data presentation and processing in a scientific context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

AsUp

-0.59

SharedDtor

-0.56

tvguidetime

-0.56

interopRequire

-0.53

 pinulongan

-0.52

oneofs

-0.49

⟬

-0.48

 nahilalakip

-0.47

 onBind

-0.46

 fédé

-0.45

POSITIVE LOGITS

 externi

0.44



0.36

wth

0.36

uero

0.36

ale

0.36

 Einzelnen

0.35

er

0.35

|')

0.35

 Staub

0.34

drücke

0.34

Activations Density 1.528%

terms and phrases related to data presentation and processing in a scientific context

No Comments

No Known Activations