INDEX

Explanations

syntax and names of programming languages or other code

oai_token-act-pair · gemini-2.0-flash

Code snippets

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.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

XmlEnum

-0.47

 stoi

-0.43

RouterModule

-0.43

complexContent

-0.42

-0.40

じん

-0.40

Toast

-0.40

<bos>

-0.40

 склада

-0.40

mensa

-0.38

POSITIVE LOGITS

 houſe

0.76

 שוליים

0.75

 purpoſe

0.73

 Majefty

0.71

 Италијани

0.69

躇

0.69

 ſche

0.68

 Houſe

0.66

 bonté

0.66

 ſtate

0.65

Activations Density 2.303%

syntax and names of programming languages or other code

Code snippets

No Comments

No Known Activations