INDEX

Explanations

phrases related to usage and access to information or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

:✨

-0.69

GEBURTSDATUM

-0.56

 itſelf

-0.52

 purpoſe

-0.50

setVerticalGroup

-0.49

VIAF

-0.49

 pleaſure

-0.48

 tagħ

-0.47

 ſche

-0.47

 myſelf

-0.46

POSITIVE LOGITS

 Zuge

0.42

nahme

0.40

Legături

0.39

șit

0.38

 StatefulWidget

0.37

 acestei

0.37

 association

0.35

PyExc

0.35

 veze

0.35

 Liang

0.35

Activations Density 0.598%

phrases related to usage and access to information or services

No Comments

No Known Activations