INDEX

Explanations

mathematical terms or expressions involving variables and numbers

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

Arithmetic sequence terms

np_acts-logits-general · gemini-2.0-flash

numbers greater than one hundred

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

 onAnimation

-0.63

UNRELATED

-0.56

Portale

-0.56

PerformLayout

-0.56

 الذات

-0.55

tvguidetime

-0.54

msgTypes

-0.54

XmlIgnore

-0.54

altham

-0.51

BeginInit

-0.51

POSITIVE LOGITS

 thousand

0.88

Thousand

0.75

 Thousand

0.74

 hundred

0.71

 THOUS

0.70

 hundreds

0.69

thousand

0.67

 thousands

0.65

 Thousands

0.63

 tausend

0.62

Activations Density 0.948%

mathematical terms or expressions involving variables and numbers

Arithmetic sequence terms

numbers greater than one hundred

No Comments

No Known Activations