INDEX

Explanations

commands or prompts related to actions

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

"Take" followed by a word

np_acts-logits-general · gemini-2.0-flash

take a | take one | take away | take care | take advantage

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

MemoryWarning

-0.95

#+#

-0.84

 doubtnut

-0.82

 itſelf

-0.77

 EClass

-0.75

theless

-0.75

 CURIAM

-0.75

 cdti

-0.74

 contextLoads

-0.74

__':

-0.72

POSITIVE LOGITS

 care

1.15

take

0.88

 Care

0.87

 advantage

0.86

care

0.86

Take

0.86

TAKE

0.82

Taking

0.78

 Take

0.77

 cared

0.77

Activations Density 0.072%

commands or prompts related to actions

"Take" followed by a word

take a | take one | take away | take care | take advantage

No Comments

No Known Activations