INDEX

Explanations

references to purpose and meaning in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

 Inscrivez

-0.66

ckså

-0.64

 CanadaChoose

-0.63

 argint

-0.63

 inoxydable

-0.63

SqlDataReader

-0.62

 aéreo

-0.60

InjectAttribute

-0.59

ptăm

-0.58

 ainfi

-0.58

POSITIVE LOGITS

 purposes

0.94

 Purpose

0.94

 purpose

0.93

Purpose

0.82

purpose

0.79

 PURPOSE

0.76

purposes

0.75

PURPOSE

0.73

 Purposes

0.72

 Zweck

0.57

Activations Density 0.318%

references to purpose and meaning in various contexts

No Comments

No Known Activations