INDEX

Explanations

performance-related terminology in programming contexts

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

Code optimization and performance

np_acts-logits-general · gemini-2.0-flash

CPU performance and optimization

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

posites

-0.50

 découvre

-0.49

 smtplib

-0.48

 détruit

-0.48

 lawfully

-0.45

 действует

-0.45

 creía

-0.45

 decidido

-0.45

 insegna

-0.45

 légal

-0.45

POSITIVE LOGITS

CPU

0.98

 performance

0.95

 load

0.87

 speed

0.83

 computationally

0.83

performance

0.82

 optimizations

0.82

 slowdown

0.80

 optimization

0.80

CPU

0.80

Activations Density 1.033%

performance-related terminology in programming contexts

Code optimization and performance

CPU performance and optimization

No Comments

No Known Activations