INDEX

Explanations

terms related to medical and biological processes or studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 olmayan

-0.28

postmedia

-0.26

ียว

-0.25

 bagay

-0.24

一種

-0.24

setCancelable

-0.24

 kesin

-0.23

 respetar

-0.23

 reactie

-0.23

Geographie

-0.22

POSITIVE LOGITS

 مشين

0.94

DoubleQuotes

0.88

 <=",

0.83

 GenerationType

0.73

))),

0.73

 виправивши

0.73

 Paglinawan

0.72

">)</

0.72

 <>",

0.71

">(</

0.71

Activations Density 2.518%

terms related to medical and biological processes or studies

No Comments

No Known Activations