INDEX

Explanations

sections of text that contain no significant content or activations

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

Tokens after dates or numbers

np_acts-logits-general · gemini-2.0-flash

dates and days

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

ectoria

-0.48

 võimal

-0.47

ništvo

-0.47

 comuniques

-0.45

 traer

-0.45

❉

-0.44

 épis

-0.44

 Urqu

-0.43

ayuno

-0.43

 verrez

-0.43

POSITIVE LOGITS

 lenker

0.76

 autorytatywna

0.70

 EconPapers

0.67

MemoryWarning

0.67

 محفوظة

0.65

typeorm

0.65

 propOrder

0.63

 nawr

0.62

Autoritní

0.62

期刊论文

0.61

Activations Density 0.185%

sections of text that contain no significant content or activations

Tokens after dates or numbers

dates and days

No Comments

No Known Activations