INDEX

Explanations

references to past events or actions that invoke regret

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

after "was" or "were"

np_acts-logits-general · gemini-2.0-flash

time of day and events

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

-0.49

Inoltre

-0.47

 lumi

-0.43

SAC

-0.43

ennung

-0.42

BUT

-0.42

illah

-0.41

 словарь

-0.41

пок

-0.41

ỏi

-0.41

POSITIVE LOGITS

 ExecuteAsync

0.71

 disambiguazione

0.67

ViewFeatures

0.64

makeConstraints

0.62

MigrationBuilder

0.62

 Roskov

0.60

 afternoon

0.60

enterOuterAlt

0.58

 specialises

0.57

ppelin

0.56

Activations Density 0.099%

references to past events or actions that invoke regret

after "was" or "were"

time of day and events

No Comments

No Known Activations