INDEX

Explanations

locations or contexts for actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 prestaciones

-0.43

 évêque

-0.37

leistung

-0.34

 Schulter

-0.33

 Gegend

-0.33

 ayudas

-0.32

 Gemeinden

-0.32

 faudrait

-0.32

 modernes

-0.31

 argentino

-0.31

POSITIVE LOGITS

ConstraintMaker

0.87

rungsseite

0.80

存于互联网档案馆

0.77

"=>$

0.76

Hochspringen

0.71

'=>$

0.70

 arşivlendi

0.69

 يتيمه

0.64

<tfoot>

0.63

UniformLocation

0.60

Activations Density 0.311%

locations or contexts for actions or events

No Comments

No Known Activations