INDEX

Explanations

references to studies, publications, and research findings related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ArgsConstructor

-0.72

下载附件

-0.67

 twofold

-0.65

 enige

-0.59

BeginInit

-0.59

 twice

-0.59

 slightly

-0.58

 noget

-0.58

 sepenuhnya

-0.57

gì

-0.57

POSITIVE LOGITS

Including

0.91

 including

0.88

including

0.88

 Including

0.86

 INCLUDING

0.78

INCLUDING

0.76

 besides

0.69

 throughout

0.67

 включая

0.66

throughout

0.65

Activations Density 0.503%

references to studies, publications, and research findings related to various topics

No Comments

No Known Activations