INDEX

Explanations

This neuron appears to activate on code, likely identifying include statements, android layout XML attributes, errors referencing files in a directory structure, and possibly chemical formulas or version numbers of programming languages.

oai_token-act-pair · gemini-2.0-flash

programming code

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_6/width_16k/average_l0_95

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.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

ScopeManager

-0.76

IBOutlet

-0.74

SceneManagement

-0.74

-0.71

lapsingToolbar

-0.68

Bibliograf

-0.67

 الحره

-0.66

kloped

-0.62

AnchorStyles

-0.61

StoryboardSegue

-0.58

POSITIVE LOGITS

 سكانية

0.60

Obrázky

0.52

 Amal

0.42

Pou

0.38

 Blaze

0.37

 Borne

0.37

 CONNECT

0.36

олові

0.36

 createContext

0.35

كتشاف

0.35

Activations Density 6.663%

This neuron appears to activate on code, likely identifying include statements, android layout XML attributes, errors referencing files in a directory structure, and possibly chemical formulas or version numbers of programming languages.

programming code

No Comments

No Known Activations

This neuron appears to activate on code, likely identifying include statements, android layout XML attributes, errors referencing files in a directory structure, and possibly chemical formulas or version numbers of programming languages.

programming code

No Comments

No Known Activations