INDEX

Explanations

names of individuals and legal terminology related to 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 @ 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

 saken

-0.83

Datuak

-0.81

sb

-0.76

 Hobson

-0.76

 TableColumn

-0.75

 Zeichen

-0.75

 Arre

-0.74

sah

-0.73

Sah

-0.72

 Ceres

-0.71

POSITIVE LOGITS

ainville

0.77

eric

0.76

 Garnier

0.75

ERICK

0.75

ніципа

0.75

 Browne

0.74

 Marquez

0.74

 Dessa

0.72

țion

0.71

 Coleridge

0.71

Activations Density 2.182%

names of individuals and legal terminology related to court cases

No Comments

No Known Activations