INDEX

Explanations

formatted and organized data structures or representations

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

<?

-0.38

 Darío

-0.37

 ligiloj

-0.36

 subterrá

-0.35

pr

-0.34

tissement

-0.34

 Widerspruch

-0.34

 botines

-0.32

øs

-0.32

 hipótesis

-0.32

POSITIVE LOGITS

 nakalista

0.65

 חיצוניים

0.61

Халык

0.60

+#+#

0.56

richTextPanel

0.55

 ddelwed

0.53

 kasarigan

0.52

SOUNDBITE

0.52

homonymie

0.52

 noDo

0.51

Activations Density 0.662%

formatted and organized data structures or representations

No Comments

No Known Activations