INDEX

Explanations

technical terms related to electrical processes and components

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

before a passive verb

np_acts-logits-general · gemini-2.0-flash

flow and spreading

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

妥

-0.52

lež

-0.49

verif

-0.48

ílio

-0.47

 Manually

-0.47

 undisclosed

-0.46

Conne

-0.46

atology

-0.45

five

-0.45

 benchmarks

-0.45

POSITIVE LOGITS

>//

0.78

 gevolg

0.68

 préfé

0.68

 يتيمه

0.67

AccessorTable

0.66

 contact

0.63

 equilibrium

0.62

 infolge

0.61

AxisAlignment

0.61



0.61

Activations Density 2.108%

technical terms related to electrical processes and components

before a passive verb

flow and spreading

No Comments

No Known Activations