INDEX

Explanations

phrases regarding advice and guidance in various contexts

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

preceding "advice" or similar terms

np_acts-logits-general · gemini-2.0-flash

advice and permission

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

odorant

-0.55

matchCondition

-0.53

 Entries

-0.49

 Marlow

-0.49

 ProtoMessage

-0.47

comand

-0.47

phosa

-0.47

rawDesc

-0.46

 Baillargeon

-0.45

 watermark

-0.45

POSITIVE LOGITS

 provided

1.45

 given

1.44

given

1.30

provided

1.25

 Provided

1.21

Provided

1.20

 received

1.20

 diberikan

1.19

 offered

1.16

Given

1.12

Activations Density 0.463%

phrases regarding advice and guidance in various contexts

preceding "advice" or similar terms

advice and permission

No Comments

No Known Activations