INDEX

Explanations

elements related to mathematical or scientific notation in the text

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

images or figures labeled (a), (b), (c)

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

 Monfieur

-0.48

recated

-0.44

 Inscrivez

-0.43

 Houſe

-0.41

iſen

-0.41

 indietro

-0.40

MFC

-0.40

拭

-0.39

Selfer

-0.38

MediaStore

-0.37

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.61

httphttps

0.50

 Мексичка

0.49

abestanden

0.45

 SwitchCompat

0.44

DataURL

0.43

 onOptions

0.42

 defaultstate

0.42

 Савезне

0.41

 sesi

0.41

Activations Density 0.038%

elements related to mathematical or scientific notation in the text

images or figures labeled (a), (b), (c)

No Comments

No Known Activations