INDEX

Explanations

references to manufacturers or protocols in scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zelle

-0.70

конец

-0.64

ITHUB

-0.63

barrow

-0.63

võ

-0.62

closeConnection

-0.61

Външни

-0.61

 noDo

-0.61

 referrerpolicy

-0.60

ضايا

-0.60

POSITIVE LOGITS

according

1.31

 according

1.29

 selon

1.26

Según

1.24

 Secondo

1.23

 לפי

1.22

 según

1.21

 secondo

1.20

 accordance

1.20

 Según

1.19

Activations Density 0.203%

references to manufacturers or protocols in scientific contexts

No Comments

No Known Activations