INDEX

Explanations

names and references related to individuals and their historical actions or statuses

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

Personensuche

-0.53

 soon

-0.37

})`

-0.35

 prochaines

-0.35

来不及

-0.33

Soon

-0.33

Ҭ

-0.33

ucap

-0.33

 prochains

-0.32

 prochain

-0.31

POSITIVE LOGITS

 modern

0.66

modern

0.62

 modernas

0.61

moderne

0.59

现代

0.58

 modernes

0.58

 moderne

0.58

 modernen

0.56

EndContext

0.55

GOTREF

0.54

Activations Density 1.009%

names and references related to individuals and their historical actions or statuses

No Comments

No Known Activations