INDEX

Explanations

references to unresolved questions and uncertainties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

เต

-0.40

 sosial

-0.38

 others

-0.37

 Sosial

-0.36

others

-0.36

 sozial

-0.36

 listos

-0.36

 today

-0.36

Nazionale

-0.35

typical

-0.35

POSITIVE LOGITS

ChildScrollView

0.68

PMailer

0.61

 كومونز

0.61

Tazama

0.58

 kasarigan

0.58

 دیکھیے

0.57

enumii

0.56

adaptiveStyles

0.56

AndEndTag

0.56

 CanadaChoose

0.56

Activations Density 0.414%

references to unresolved questions and uncertainties

No Comments

No Known Activations