INDEX

Explanations

numerical values related to ranking or categorization

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

top

-0.57

ArgsConstructor

-0.48

Top

-0.48

parsedMessage

-0.47

BufferException

-0.46

czo

-0.45

 Wicidata

-0.43

########.

-0.43

TOP

-0.42

MathML

-0.42

POSITIVE LOGITS

AnimationsModule

0.49

المكان

0.42

 незавершена

0.39

Initializable

0.37

 verwijzen

0.37

ModelSerializer

0.35

DeclareMath

0.35

 StatefulWidget

0.35

NewUrlParser

0.35

 виправивши

0.34

Activations Density 0.003%

numerical values related to ranking or categorization

No Known Activations