INDEX

Explanations

mathematical terms and concepts related to Taylor expansions and derivatives

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

Taylor series expansion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

extrême

-0.48

 dieß

-0.44

ſelves

-0.42

équi

-0.42

 Continent

-0.41

 Geſ

-0.41

 enfans

-0.40

posizioni

-0.40

 élect

-0.40

posals

-0.40

POSITIVE LOGITS

Personendaten

0.59

 pinulongan

0.45

IndentedString

0.43

جغرافيا

0.42

 <<<<<<<<<<<<<<

0.42

 ExecuteAsync

0.41

AndEndTag

0.40

uginosa

0.40

 ویکی‌پدیا

0.40

ResumeLayout

0.39

Activations Density 0.127%

mathematical terms and concepts related to Taylor expansions and derivatives

Taylor series expansion

No Comments

No Known Activations