INDEX

Explanations

terms related to gratitude, appreciation, and praise

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

of gratitude or commendation

np_acts-logits-general · gemini-2.0-flash

praise and appreciation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 proving

-0.62

PullParser

-0.58

تأ

-0.55

.*")]

-0.55

 Proving

-0.53

weile

-0.52

lixir

-0.52

出版年

-0.51

зумі

-0.50

urlpatterns

-0.50

POSITIVE LOGITS

worthy

0.77

ably

0.72

fully

0.71

EDEFAULT

0.71

 רבה

0.64

FULLY

0.62

받

0.61

ьаж

0.61

नीय

0.60

tably

0.56

Activations Density 0.213%

terms related to gratitude, appreciation, and praise

of gratitude or commendation

praise and appreciation

No Comments

No Known Activations