INDEX

Explanations

variations and applications of specific verbs and phrases related to usage and implementation

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

placing or using something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

WriteTagHelper

-0.63

 contextLoads

-0.58

MemoryWarning

-0.57

RemoveField

-0.56

 الرياضيه

-0.54

 mijne

-0.53

 препратки

-0.52

 eenig

-0.52

 enfans

-0.51

 avoient

-0.51

POSITIVE LOGITS

󠁿

0.40

 usable

0.39

 tote

0.38

gra

0.38

pot

0.37

pod

0.36

 hopefully

0.36

tagHelperRunner

0.35

usable

0.35

Pot

0.35

Activations Density 0.047%

variations and applications of specific verbs and phrases related to usage and implementation

placing or using something

No Comments

No Known Activations