INDEX

Explanations

symbols and formatting related to data structure or documentation formats

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

Figure/panel labels in scientific papers

np_acts-logits-general · gemini-2.0-flash

list items and diagrams

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

########.

-0.68

 typelib

-0.60

 snippetHide

-0.59

WebVitals

-0.54

MigrationBuilder

-0.52

 disambiguazione

-0.52

VersionUID

-0.51

 ***!

-0.47

autoreleasepool

-0.47

原始内容存档于

-0.44

POSITIVE LOGITS

 pleaſure

0.40

 Diſ

0.39

 Perſ

0.39

blatt

0.38

writerow

0.38

updateUser

0.37

 Kapit

0.37

ünstler

0.36

Escape

0.35

addComponent

0.35

Activations Density 0.056%

symbols and formatting related to data structure or documentation formats

Figure/panel labels in scientific papers

list items and diagrams

No Comments

No Known Activations