INDEX

Explanations

references to technology and data-driven analysis in various contexts

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

Preceding words related to purpose or intention

np_acts-logits-general · gemini-2.0-flash

used to define purpose

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

tilhe

-0.64

TO

-0.58

RunWith

-0.57

AtIndexPath

-0.56

Her

-0.55

GeneratedMessage

-0.55

ocarpus

-0.55

AutoField

-0.55

eriks

-0.53

readValue

-0.53

POSITIVE LOGITS

to

1.46

 чтобы

1.33

 để

1.28

 upang

1.22

 כדי

1.20

 щоб

1.09

 untuk

0.94

来

0.94

เพื่อ

0.92

 чтоб

0.89

Activations Density 1.211%

references to technology and data-driven analysis in various contexts

Preceding words related to purpose or intention

used to define purpose

No Comments

No Known Activations