INDEX

Explanations

references to annual events and reports

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

annual + occurrence/event/report

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

abet

-0.47

bey

-0.47

 herencia

-0.43

 cozinha

-0.43

 maksud

-0.42

Forsch

-0.42

 Ciencias

-0.42

 Gegend

-0.42

 Cientí

-0.41

Seb

-0.41

POSITIVE LOGITS

 annual

0.99

Annual

0.98

 Annual

0.96

annual

0.91

 ANNUAL

0.88

 Yearly

0.85

ANNUAL

0.83

 anual

0.83

 yearly

0.81

Yearly

0.79

Activations Density 0.008%

references to annual events and reports

annual + occurrence/event/report

No Comments

No Known Activations