INDEX

Explanations

terms related to scientific methods and principles

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 pinulongan

-0.68

rungsseite

-0.66

:✨

-0.60

MenuInflater

-0.59

 EconPapers

-0.57

脚注の使い方

-0.56

 Chwiliwch

-0.54



-0.54

orteur

-0.51

iredo

-0.50

POSITIVE LOGITS

原理

0.60

 concept

0.54

Princip

0.52

 Princip

0.51

 principle

0.50

concept

0.48

 PRINCIP

0.46

PRINCIP

0.46

princip

0.46

 Concept

0.46

Activations Density 1.222%

terms related to scientific methods and principles

No Known Activations