INDEX

Explanations

frequent articles and conjunctions in text

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

 समीक्षाएं

-0.63

 CURIAM

-0.62

 myſelf

-0.61

 itſelf

-0.60

verifyException

-0.58

WriteBarrier

-0.57

bootstrapcdn

-0.57

expandindo

-0.56

:✨

-0.56

yarnpkg

-0.56

POSITIVE LOGITS

0.33

pa

0.32

0.31

whether

0.31

gantung

0.30

lie

0.30

ž

0.30

0.29

sob

0.29

Activations Density 0.012%

frequent articles and conjunctions in text

No Comments

No Known Activations