INDEX

Explanations

coding syntax and programming elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Code snippets: paths/filenames

np_acts-logits-general · gemini-2.0-flash

code definitions and identifiers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ſelves

-0.78

ſelf

-0.71

 pleaſure

-0.68

 ſche

-0.67

 ſur

-0.66

 houſe

-0.65

 queſta

-0.63

:✨

-0.63

 tranſ

-0.62

 itſelf

-0.62

POSITIVE LOGITS

TabIndex

0.49

endregion

0.42

 Schroeder

0.40

 Contreras

0.40

 Winkler

0.39

PrivateRoute

0.38

 Stolz

0.37

jsxFileName

0.35

 Conferencia

0.35

Activations Density 0.850%

coding syntax and programming elements

Code snippets: paths/filenames

code definitions and identifiers

No Comments

No Known Activations