INDEX

Explanations

references to hierarchical or noble titles and their associated themes

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

ViewImports

-0.51

MemoryWarning

-0.50

AndEndTag

-0.50

ftagPool

-0.49

featureID

-0.48

 GreatSchools

-0.45

şört

-0.43

InstrumentedTest

-0.43

 StringTokenizer

-0.42

хьтан

-0.41

POSITIVE LOGITS

 مشين

0.46

Pyx

0.43

ۜ

0.41

likle

0.40

MergeFrom

0.39

 kasarigan

0.38

帯

0.38

.-(

0.38

 հղումներ

0.38

🏻

0.37

Activations Density 0.277%

references to hierarchical or noble titles and their associated themes

No Comments

No Known Activations