INDEX

Explanations

structured numerical data and statistical references

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

citations and year numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

Халык

-0.63

Бахар

-0.61

Portail

-0.60

Демографія

-0.57

 autorytatywna

-0.56

qrstuvwxyz

-0.55

TagMode

-0.55

 للمعارف

-0.52

لينكات

-0.52

oa̍t

-0.51

POSITIVE LOGITS

 Dijo

0.43

ValueStyle

0.41

 actualité

0.41

 asimismo

0.40

 misma

0.40

 mismo

0.39

 delegado

0.39

 also

0.39

 detenido

0.38

 aprobado

0.38

Activations Density 0.061%

structured numerical data and statistical references

citations and year numbers

No Comments

No Known Activations