INDEX

Explanations

instances of the word "fallen" and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

 Städ

-0.47

不及

-0.41

ฉ

-0.39

 pharmaco

-0.39

}\,

-0.38

 Nowak

-0.38

 mandated

-0.38

 gesetz

-0.37

 avut

-0.37

 Carrasco

-0.37

POSITIVE LOGITS

 dropped

0.89

 dropping

0.83

Drop

0.78

 Drop

0.76

Dro

0.72

drop

0.71

Dro

0.71

 betweenstory

0.71

 drop

0.70

dropping

0.68

Activations Density 0.260%

instances of the word "fallen" and its variants

No Comments

No Known Activations