INDEX

Explanations

content related to court rulings and judicial errors

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

court actions

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

Portail

-0.39

-0.34

 înal

-0.33

исленность

-0.33

 utafitiHapana

-0.32

»

-0.32

➩

-0.32

pile

-0.31

”

-0.31

¸

-0.31

POSITIVE LOGITS

 CURIAM

0.75

SequentialGroup

0.66

 ddelwed

0.59

EndInit

0.59

ftagPool

0.58

новништво

0.57

 dieſes

0.55

]),

0.55

beiros

0.54

 ſehen

0.54

Activations Density 0.142%

content related to court rulings and judicial errors

court actions

No Comments

No Known Activations