INDEX

Explanations

definitions and characteristics of concepts or phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 themselves

-0.41

 którzy

-0.41

 themſelves

-0.41

 wuß

-0.39

 kteří

-0.38

themselves

-0.37

 myſelf

-0.36

 hacían

-0.36

 weren

-0.34

 olduk

-0.33

POSITIVE LOGITS

Personendaten

0.71

AndEndTag

0.68

Autoritní

0.65

 مرئيه

0.65

rungsseite

0.60

 unites

0.60

ValueStyle

0.59

<bos>

0.59

delwed

0.57

PyExc

0.57

Activations Density 0.987%

definitions and characteristics of concepts or phenomena

No Comments

No Known Activations