INDEX

Explanations

legal terminology and references related to court proceedings and testimonies

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

claim or accusation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 hiszen

-0.42

gól

-0.42

 CURIAM

-0.42

Literals

-0.40

 anse

-0.40

 мәкал

-0.40

alisa

-0.38

CF

-0.37

είο

-0.37

だな

-0.37

POSITIVE LOGITS

 allegedly

3.09

 supposedly

2.92

 supuestamente

2.63

 alleged

2.61

 purported

2.47

supposed

2.36

 reportedly

2.27

 supposed

2.25

 angeb

2.08

alleg

2.08

Activations Density 0.733%

legal terminology and references related to court proceedings and testimonies

claim or accusation

No Comments

No Known Activations