INDEX

Explanations

references to legal instructions and evidence in court cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 Schreiben

-0.63

hynchus

-0.61

leby

-0.57

IRUS

-0.52

});*/

-0.52

Rohy

-0.52

 bureaucrats

-0.50

 Outlet

-0.50

deleteById

-0.50

angliski

-0.49

POSITIVE LOGITS

 crossorigin

0.59

$_['

0.58

řevě

0.56

 BrowserModule

0.56

JUR

0.55

Vidite

0.54

/**

0.54

 propOrder

0.52

 fVar

0.50

 jury

0.50

Activations Density 1.209%

references to legal instructions and evidence in court cases

No Comments

No Known Activations