INDEX

Explanations

code or programming-related structures and comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 للاسماء

-0.62

 casó

-0.59

 mérit

-0.50

 vectoriales

-0.47

UIControlState

-0.45

 dévelo

-0.43

 translateY

-0.43

 Broward

-0.41

 Abonnez

-0.41

 feroit

-0.40

POSITIVE LOGITS

Portale

0.86

RegressionTest

0.81

Personendaten

0.71

بوابة

0.71

Portail

0.70

 ErrIntOverflow

0.68

usermodel

0.67

setVerticalGroup

0.67

0.66

原始内容存档于

0.66

Activations Density 0.251%

code or programming-related structures and comments

No Comments

No Known Activations