INDEX

Explanations

programming comments and preprocessor directives in code

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

conditional compilation directives

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

 curt

-0.42

Css

-0.41

 poetry

-0.41

 narr

-0.40

OSM

-0.40

 hara

-0.40

CreateMap

-0.40

 sorcerer

-0.40

 lore

-0.40

 Lorentz

-0.39

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.63

GEBURTS

0.63

fjspx

0.59

endif

0.59

httphttps

0.55

pexpr

0.55

 endif

0.50

indd

0.50

WriteAttribute

0.49

 Factbook

0.48

Activations Density 0.013%

programming comments and preprocessor directives in code

conditional compilation directives

No Comments

No Known Activations