INDEX

Explanations

phrases indicating a high quantity or number

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 approximately

-0.62

approximately

-0.62

Approximately

-0.60

 Approximately

-0.56

GEBURTSDATUM

-0.56

 aproximadamente

-0.56

 Roskov

-0.56

 approx

-0.52

 Approx

-0.49

Approx

-0.48

POSITIVE LOGITS

 dozen

0.63

 half

0.60

 halfway

0.57

 hundred

0.54

dozen

0.53

 EconPapers

0.45

 HALF

0.45

百

0.45

 hundert

0.44

 деся

0.43

Activations Density 0.245%

phrases indicating a high quantity or number

No Comments

No Known Activations