INDEX

Explanations

phrases indicating strong recommendations or must-see content

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

"must" or "worth" followed by an action

np_acts-logits-general · gemini-2.0-flash

must read or watch

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

ukone

-0.68

-0.67

transQ

-0.60

MemoryWarning

-0.60

 Мексичка

-0.59

IUrlHelper

-0.59

DrawerToggle

-0.58

MigrationBuilder

-0.58

 AssemblyProduct

-0.56

webElement

-0.55

POSITIVE LOGITS

 MUST

0.98

must

0.94

 must

0.94

 worth

0.93

 definitely

0.92

Must

0.88

 unbedingt

0.86

 Must

0.86

Worth

0.86

Definitely

0.84

Activations Density 0.129%

phrases indicating strong recommendations or must-see content

"must" or "worth" followed by an action

must read or watch

No Comments

No Known Activations