INDEX

Explanations

the end of code blocks and conditional statements in various programming languages, especially those using curly braces, parentheses, and square brackets

oai_token-act-pair · gemini-2.0-flash

}

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_23/width_16k/average_l0_25

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 propOrder

-1.31

 للاسماء

-1.31

parsedMessage

-1.29

Personendaten

-1.26

 تضيفلها

-1.22

rungsseite

-1.18

 <=",

-1.16

PhysRevD

-1.15

цездатний

-1.13

 '\\;'

-1.12

POSITIVE LOGITS

0.51

0.49

↵

0.46

if

0.46

↵↵

0.41

0.40

all

0.39

0.38

<eos>

0.38

Activations Density 1.075%

the end of code blocks and conditional statements in various programming languages, especially those using curly braces, parentheses, and square brackets

}

No Comments

No Known Activations

the end of code blocks and conditional statements in various programming languages, especially those using curly braces, parentheses, and square brackets

}

No Comments

No Known Activations