INDEX

Explanations

terms related to research studies and analysis

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 justiça

-0.57

 elástica

-0.56

cipar

-0.56

 Dispens

-0.53

 sementes

-0.53

 operacional

-0.53

 proudest

-0.53

ंदीखरीदारी

-0.53

 fantasma

-0.53

 Rass

-0.52

POSITIVE LOGITS

 study

1.40

 studied

1.34

 examine

1.28

 investigate

1.28

 examining

1.27

 studying

1.27

 examined

1.26

 investigated

1.25

 examines

1.23

studied

1.22

Activations Density 0.565%

terms related to research studies and analysis

No Known Activations