INDEX

Explanations

phrases indicating recommendations and suitability for specific audiences or interests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 équ

-0.37

 deviennent

-0.35

 spille

-0.33

 ribuan

-0.31

 dział

-0.30

설

-0.30

"]]

-0.30

 recev

-0.30

 medzi

-0.29

 beziehen

-0.28

POSITIVE LOGITS

ValueStyle

0.73

 protoimpl

0.69

expandindo

0.69

EndGlobalSection

0.68

 unknownFields

0.62

 يتيمه

0.60

FormTagHelper

0.60

 autorytatywna

0.59

awtextra

0.59

styleType

0.58

Activations Density 0.141%

phrases indicating recommendations and suitability for specific audiences or interests

No Comments

No Known Activations