INDEX

Explanations

keywords and phrases related to data management and user actions within a system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

<eos>

-0.64

Air

-0.54

“

-0.51

 […]

-0.51

air

-0.51

 land

-0.48

Th

-0.48

The

-0.47

se

-0.46

Se

-0.46

POSITIVE LOGITS

 uſed

1.01

 Jefus

1.00

 poffe

0.96

 HasFactory

0.95

 myſelf

0.94

 poffible

0.88

 leſs

0.87

 незавершена

0.87

 ſeveral

0.86

 Diretto

0.86

Activations Density 0.411%

keywords and phrases related to data management and user actions within a system

No Comments

No Known Activations