INDEX

Explanations

programming-related documentation and annotations

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

equation labels

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

 Infórmanos

-0.71

ImageContext

-0.66

XmlAccessType

-0.63

ConstraintMaker

-0.60

/**

-0.60

 Signalez

-0.59

 GenerationType

-0.59

 Italijani

-0.58

TestingModule

-0.57

⤹

-0.57

POSITIVE LOGITS

 sindaco

0.32

avras

0.31

opi

0.30

 marito

0.29

webElement

0.29

slideUp

0.29

 Mots

0.29

 Wolf

0.28

 magnético

0.28

 opio

0.28

Activations Density 0.003%

programming-related documentation and annotations

equation labels

No Comments

No Known Activations