INDEX

Explanations

code snippets, especially in languages like PHP and possibly Portuguese (based on words like "seu" and "escreva").

oai_token-act-pair · gemini-2.0-flash

mixed foreign language and code

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

CreateModel

-0.46

 send

-0.40

Kariera

-0.39

 getSource

-0.38

 محفوظة

-0.37

 MainAxisSize

-0.37

HasAnnotation

-0.37

 basic

-0.37

 свят

-0.36

 spielt

-0.36

POSITIVE LOGITS

Efq

0.91

 ſche

0.83

 ſeveral

0.81

 Theſe

0.79

 itſelf

0.77

 myſelf

0.76

 themſelves

0.75

 houſe

0.75

 Houſe

0.73

 ſtate

0.73

Activations Density 0.706%

code snippets, especially in languages like PHP and possibly Portuguese (based on words like "seu" and "escreva").

mixed foreign language and code

No Comments

No Known Activations