INDEX

Explanations

expressions of gratitude

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

Personendaten

-0.42

 hemel

-0.42

 ethical

-0.42

protoimpl

-0.41

 péné

-0.40

 wiers

-0.39

 abito

-0.38

fficio

-0.38

fileID

-0.38

愛知県

-0.38

POSITIVE LOGITS

Thanks

0.90

thanks

0.89

 Thanks

0.89

THANKS

0.86

 thanks

0.85

 THANKS

0.82

Thx

0.79

Thx

0.73

Thanx

0.72

thx

0.68

Activations Density 0.086%

expressions of gratitude

No Comments

No Known Activations