INDEX

Explanations

annotations and metadata in code, particularly those used for persistence and serialization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 betweenstory

-0.82

 MonoBehaviour

-0.78

 препратки

-0.77

Spoljašnje

-0.77

 BorderRadius

-0.75

AsUp

-0.72

 للمعارف

-0.70

şiv

-0.69

 tartalomajánló

-0.69

unmodifiable

-0.66

POSITIVE LOGITS

SerializedName

0.57

 Doyle

0.51

RequestMapping

0.50

期刊论文

0.49

 RequestMethod

0.49

 quran

0.49

JsonProperty

0.48

JoinColumn

0.48

英語版

0.47

ærke

0.47

Activations Density 0.042%

annotations and metadata in code, particularly those used for persistence and serialization

No Comments

No Known Activations