INDEX

Explanations

phrases indicating disagreement or urging caution in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

MethodManager

-0.50

bootstrapcdn

-0.48

enschappelijke

-0.41

 Ressources

-0.41

 SwitchCompat

-0.41

的呢

-0.40

 bougies

-0.40

<()>

-0.39

Objective

-0.39

 intStringLen

-0.39

POSITIVE LOGITS

 jangan

0.51

 Jangan

0.50

 Remember

0.50

 đừng

0.49

Jangan

0.48

be

0.48

 不要

0.46

don

0.45

Don

0.45

 considérons

0.44

Activations Density 0.190%

phrases indicating disagreement or urging caution in discussions

No Comments

No Known Activations