INDEX

Explanations

technical or scientific terminology related to data analysis and modeling

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

modeling, vertex, scanning, tuple

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

 information

-0.26

 added

-0.25

ice

-0.23

 Information

-0.22

 afectan

-0.21

 effect

-0.20

 daarom

-0.20

↵↵

-0.20

恩

-0.20

 rate

-0.20

POSITIVE LOGITS

OGND

1.08

 CreateTagHelper

0.95

 pinulongan

0.92

 SwitchCompat

0.87

verifyException

0.85

 ویکی‌پدی

0.82

 незавершена

0.82

<unused16>

0.82

<unused3>

0.82

<unused8>

0.81

Activations Density 0.137%

technical or scientific terminology related to data analysis and modeling

modeling, vertex, scanning, tuple

No Comments

No Known Activations