INDEX

Explanations

legal terminology and names related to court cases or judicial opinions

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

legal phrases and code keywords

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

 Marín

-0.36

 Haller

-0.32

Day

-0.31

 Schult

-0.30

 pouvoir

-0.30

ers

-0.30

RemoveAt

-0.29

 movement

-0.29

POSITIVE LOGITS

openg

1.14

quiao

1.00

 CURIAM

0.99

Hentet

0.98

 NSError

0.79

gettext

0.79

parsedMessage

0.73

 objectMapper

0.71

epa

0.70

 Phal

0.70

Activations Density 0.007%

legal terminology and names related to court cases or judicial opinions

legal phrases and code keywords

No Comments

No Known Activations