INDEX

Explanations

code snippets with semicolons, parenthesis, return commands, and the word "this"

oai_token-act-pair · gemini-2.0-flash

code syntax elements and return statements

np_token-act-pair-logits · gpt-4o-mini

semicolon/parenthesis

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 caccia

-0.50

-0.47

sir

-0.47

-0.46

 păr

-0.45

mdl

-0.44

-0.43

ługa

-0.43

 masina

-0.43

POSITIVE LOGITS

NameInMap

0.89

 الرياضيه

0.84

AddTagHelper

0.84

RenderAtEndOf

0.82

Geplaatst

0.82

 ligiloj

0.81

contentLoaded

0.80

 ujednoznacz

0.79

 ostavi

0.78

 propOrder

0.78

Activations Density 2.166%

code snippets with semicolons, parenthesis, return commands, and the word "this"

code syntax elements and return statements

semicolon/parenthesis

No Comments

No Known Activations