INDEX

Explanations

numeric or alphanumeric identifiers and addresses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 okay

-1.67

ok

-1.63

OK

-1.62

 alright

-1.37

 OKAY

-1.31

OK

-1.29

 Okay

-1.27

okay

-1.23

Okay

-1.17

ok

-1.16

POSITIVE LOGITS

Personendaten

0.76

🏾

0.55

cellular

0.55

NUMX

0.55

^(@)

0.54

 Wikimedijinoj

0.54

xsi

0.54

__':

0.53

phosa

0.53

 nakalista

0.53

Activations Density 0.081%

numeric or alphanumeric identifiers and addresses

No Comments

No Known Activations