INDEX

Explanations

references to legal cases and judicial terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Keane

-0.99

 Marne

-0.94

 hemispheres

-0.90

aians

-0.86

voorbeeld

-0.86

rsiniz

-0.85

 Infórmanos

-0.84

 Paine

-0.83

 Huss

-0.83

Hearts

-0.82

POSITIVE LOGITS

ISupport

0.85

0.73

Prés

0.72

 Mustangs

0.71

Wyatt

0.71

CodeAttribute

0.71

jen

0.71

Jeb

0.70

Rosie

0.70

 Luch

0.69

Activations Density 2.198%

references to legal cases and judicial terminology

No Comments

No Known Activations