INDEX

Explanations

syntax related to mathematical expressions, programming structures, and error handling

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

aligned code and math

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 незавершена

-1.31

WebVitals

-0.94

GEBURTSDATUM

-0.84

 estekak

-0.81

.*")]

-0.77

IntoConstraints

-0.76

 disambiguazione

-0.76

 ProtoMessage

-0.75

writeFieldEnd

-0.73

ImageContext

-0.73

POSITIVE LOGITS

Blue

0.38

Similar

0.37

 Similar

0.35

 Blue

0.34

 similar

0.32

Emp

0.32

SAV

0.32

Long

0.32

 affidavit

0.31

emp

0.31

Activations Density 0.038%

syntax related to mathematical expressions, programming structures, and error handling

aligned code and math

No Comments

No Known Activations