INDEX

Explanations

expressions of regret or pleas for forgiveness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 !!!!!!

-0.98

 !!!!

-0.98

!!!

-0.94

 !!!!!

-0.93

!!

-0.85

..!!

-0.72

！！！！

-0.71

¡¡¡

-0.71

！！！！！

-0.71

！！！

-0.71

POSITIVE LOGITS

0.75

0.58

!”

0.44

essentiel

0.39

!"

0.39

 znajdziesz

0.39

 अलावा

0.38

!」

0.36

 neutre

0.36

!)

0.36

Activations Density 0.188%

expressions of regret or pleas for forgiveness

No Comments

No Known Activations