INDEX

Explanations

technical processes or terms related to instructions or methodologies

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_9/width_16k/average_l0_100

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 planejamento

-0.47

SequentialGroup

-0.46

arrêter

-0.45

 dignidad

-0.45

attendre

-0.44

 zelfs

-0.44

 montaż

-0.44

 verdaderamente

-0.43

Vidite

-0.43

 bahkan

-0.43

POSITIVE LOGITS

 pinulongan

0.52

UD

0.48

 circu

0.45

BCA

0.44

ICA

0.44

DMETHOD

0.43

عرِّف

0.43

LPC

0.43

出版年

0.43

SPM

0.43

Activations Density 1.294%

technical processes or terms related to instructions or methodologies

No Comments

No Known Activations