INDEX

Explanations

phrases indicating belief, assertion, or opinion about subjects

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

Preceding verbs followed by "to"

np_acts-logits-general · gemini-2.0-flash

reported beliefs or observations

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

 /\.(

-0.51

zzar

-0.50

 esternos

-0.49

and

-0.48

flores

-0.48

<?=

-0.48

 ComVisible

-0.47

queues

-0.46

tor

-0.45

atoon

-0.44

POSITIVE LOGITS

EDEFAULT

0.72

 ότι

0.65

 BoxFit

0.64

ormais

0.62

AxisAlignment

0.61

 hefyd

0.61

 muualla

0.61

 TimeUnit

0.61

 '\\;'

0.60

Дереккөздер

0.59

Activations Density 0.398%

phrases indicating belief, assertion, or opinion about subjects

Preceding verbs followed by "to"

reported beliefs or observations

No Comments

No Known Activations