INDEX

Explanations

aspects of procedural effectiveness and comparisons in a narrative context

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

lists of items and details

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

رشف

-0.44

principalColumn

-0.43

FXMLLoader

-0.39

look

-0.37

 tires

-0.36

 Franch

-0.36

DrawerToggle

-0.36

Flyer

-0.36

原始内容存档于

-0.35

GEBURTS

-0.35

POSITIVE LOGITS

 autorytatywna

0.51

UnusedPrivate

0.47

esModule

0.46

 considérons

0.45

 متعلقه

0.44

ipelago

0.44

TRAILING

0.40

xase

0.40

 nữa

0.40

ynos

0.39

Activations Density 0.048%

aspects of procedural effectiveness and comparisons in a narrative context

lists of items and details

No Comments

No Known Activations