INDEX

Explanations

numerous occurrences of the word "a" and its variations, indicating a focus on articles in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 admittedly

-0.38

 özellikle

-0.37

 zuletzt

-0.35

 yalnızca

-0.35

ńskich

-0.35

ically

-0.34

 ligiloj

-0.34

 künftig

-0.33

 şimdi

-0.33

 ausz

-0.33

POSITIVE LOGITS

 nahilalakip

0.79

GEBURTSDATUM

0.65

出版年

0.63

 rhestr

0.62

 perfect

0.60

 Италијани

0.59

 certain

0.58

 resourceCulture

0.57

 great

0.57

zzleHttp

0.57

Activations Density 0.624%

numerous occurrences of the word "a" and its variations, indicating a focus on articles in the text

No Comments

No Known Activations