INDEX

Explanations

specific occurrences of programming-related terms or structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

TagMode

-0.81

ensement

-0.77

VersionUID

-0.71

complexType

-0.71

хьтан

-0.69

ⓧ

-0.69

queryInterface

-0.68



-0.66

ymce

-0.65

NameInMap

-0.63

POSITIVE LOGITS

 AppColors

0.59

"]/

0.55

 '/';

0.52

}^{+\

0.49

erweise

0.48

 "'");

0.48

 @"";

0.47

 "/";

0.47

 mój

0.47

τεύ

0.47

Activations Density 0.453%

specific occurrences of programming-related terms or structures

No Comments

No Known Activations