INDEX

Explanations

phrases indicating an action of returning or referencing back to previously mentioned text

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

"Return" followed by "to"

np_acts-logits-general · gemini-2.0-flash

return to

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

DeleteBehavior

-0.65

drücken

-0.58

Kens

-0.57

épisode

-0.55

 Carlsbad

-0.55

stellen

-0.55

หน้าน

-0.55

erráneo

-0.54

CharStream

-0.54

 desear

-0.54

POSITIVE LOGITS

 Returns

0.89

 returns

0.86

RETURNS

0.79

 pulumi

0.78

Returns

0.77

returns

0.75

 مشين

0.74

 विश्वसनीयता

0.73

 Return

0.73

eturn

0.73

Activations Density 0.107%

phrases indicating an action of returning or referencing back to previously mentioned text

"Return" followed by "to"

return to

No Comments

No Known Activations