INDEX

Explanations

numerical and quantitative information

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

乾

-0.41

tender

-0.39

 Enth

-0.37

 IBOutlet

-0.36

reducers

-0.36

Eg

-0.35

kac

-0.35

 teil

-0.34

++];

-0.34

emptyset

-0.33

POSITIVE LOGITS

now

0.61

 désormais

0.61

 ormai

0.60

 artık

0.59

 nunmehr

0.53

StructEnd

0.52

middels

0.52

 теперь

0.52

 ahora

0.52

 agora

0.49

Activations Density 0.499%

numerical and quantitative information

No Comments

No Known Activations