INDEX

Explanations

patterns related to low-level programming or system calls

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

Hexadecimal code representations

np_acts-logits-general · gemini-2.0-flash

hexadecimal codes and programming identifiers

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

 avoient

-0.82

 répondit

-0.78

 étoit

-0.78

 ainfi

-0.77

 étoient

-0.73

 stiefel

-0.71

 auroit

-0.70

 feroit

-0.70

 tendenza

-0.68

 enfans

-0.67

POSITIVE LOGITS

etc

0.75

 CreateTagHelper

0.69

nonatomic

0.62

Hochspringen

0.59

etc

0.59

帖最后由

0.58

 propOrder

0.57

VersionUID

0.57

cstdlib

0.57

webElementXpaths

0.56

Activations Density 0.265%

patterns related to low-level programming or system calls

Hexadecimal code representations

hexadecimal codes and programming identifiers

No Comments

No Known Activations