INDEX

Explanations

expressions of gratitude and friendliness

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

Sign-offs and closings

np_acts-logits-general · gemini-2.0-flash

Warm regards, sincerely, cheers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 للمعارف

-0.71

่านั้น

-0.64

 betweenstory

-0.63

出版年

-0.62

ragalactic

-0.62

 يتيمه

-0.61

いかがでしたか

-0.60

зулта

-0.60

UnusedPrivate

-0.59

WebServlet

-0.57

POSITIVE LOGITS

 Regards

1.25

Regards

1.24

Sincerely

1.07

 regards

1.04

regards

1.03

 Sincerely

0.97

Cheers

0.96

 Cheers

0.93

Best

0.90

 best

0.89

Activations Density 0.175%

expressions of gratitude and friendliness

Sign-offs and closings

Warm regards, sincerely, cheers

No Comments

No Known Activations