INDEX

Explanations

mentions of dates, events, and relationships involving historical figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

enumi

-0.34

 sufficient

-0.30

 enough

-0.30

 Monti

-0.29

 suficientes

-0.29

 cuantas

-0.28

 suficiente

-0.27

↵↵

-0.26

 hope

-0.26

POSITIVE LOGITS

iſchen

0.87

 vooz

0.87

 geweſen

0.85

 surla

0.85

 ujednoznacz

0.85

MLLoader

0.84

niſſe

0.83

 ſei

0.82

 verſch

0.82

ſſung

0.82

Activations Density 0.923%

mentions of dates, events, and relationships involving historical figures

No Comments

No Known Activations