INDEX

Explanations

geographic features and locations

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

 internos

-0.38

AccessFile

-0.37

 verksamhet

-0.36

bő

-0.35

AnchorStyles

-0.34

 bellezza

-0.33

 Mahasiswa

-0.33

来得

-0.33

merkungen

-0.32

 lettera

-0.31

POSITIVE LOGITS

 crossing

0.84

 crossings

0.80

Passage

0.79

 bridge

0.77

 Passage

0.77

crossing

0.76

 passage

0.75

passage

0.74

 bridges

0.73

 gate

0.73

Activations Density 0.815%

geographic features and locations

No Comments

No Known Activations