INDEX

Explanations

numerical data and specific dates related to medical or scientific observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

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

 ivelany

-0.69

grà

-0.66

Kedua

-0.66

 Kedua

-0.64

❷

-0.64

تقاوى

-0.59

 dois

-0.58

two

-0.58

 twice

-0.57

 beiden

-0.57

POSITIVE LOGITS

اول

0.56

pædia

0.55

 first

0.55

 FIRST

0.50

 First

0.49

Aly

0.48

FIRST

0.48

 primera

0.48

ראש

0.48

 Single

0.48

Activations Density 0.530%

numerical data and specific dates related to medical or scientific observations

No Comments

No Known Activations