INDEX

Explanations

references to authors and their contributions in academic or formal contexts

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

Preceding words of "author"

np_acts-logits-general · gemini-2.0-flash

author or writer

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 вперед

-0.55

::::::::

-0.51

 nawr

-0.48

centerY

-0.47

lawan

-0.47

 umana

-0.46

liptic

-0.46

You

-0.45

 femininas

-0.45

 Erben

-0.45

POSITIVE LOGITS

 author

1.77

 writer

1.60

 authors

1.45

author

1.39

 Author

1.32

 undersigned

1.25

Author

1.22

writer

1.22

 Authors

1.19

 Writer

1.16

Activations Density 0.256%

references to authors and their contributions in academic or formal contexts

Preceding words of "author"

author or writer

No Comments

No Known Activations