INDEX

Explanations

references to specific dates and statistical data

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

years starting with 199

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

 sirop

-0.42

VIAF

-0.38

SuppressLint

-0.38

 Merah

-0.36

WebVitals

-0.36

 للمعارف

-0.35

 sirup

-0.35

iedene

-0.33

Contoh

-0.32

Katso

-0.32

POSITIVE LOGITS

 OFDb

0.66

 EconPapers

0.61

 millennium

0.60

AndEndTag

0.57

gridy

0.56

ckley

0.54

sidemargin

0.54

Vidite

0.53

ennium

0.52

ASCII

0.52

Activations Density 0.060%

references to specific dates and statistical data

years starting with 199

No Comments

No Known Activations