INDEX

Explanations

references to unusual or unexpected events and actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 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

+#+#

-0.52

 présidenti

-0.47

 avancée

-0.45

 engraçadas

-0.43

 primeiras

-0.43

 éditoriale

-0.42

 écologique

-0.41

 pères

-0.41

 plantilla

-0.41

 Offisielt

-0.41

POSITIVE LOGITS

Rel

0.40

ri

0.40

 <>",

0.40

تقاوى

0.40

 bushes

0.40

 Something

0.40

 kéo

0.40

 centi

0.38

 Relief

0.38

DockStyle

0.38

Activations Density 0.333%

references to unusual or unexpected events and actions

No Comments

No Known Activations