INDEX

Explanations

references to technical details and specifications related to components and processes

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

need for subsequent additional

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

انتهای

-0.49

 CreateTagHelper

-0.48

SharedCtor

-0.45

 חיצוניים

-0.40

PositiveButton

-0.39

\}\\

-0.39

SourceChecksum

-0.38

endpush

-0.36

はじめに

-0.36

 IActionResult

-0.36

POSITIVE LOGITS

ViewFeatures

0.41

 moletom

0.40

UnusedPrivate

0.39

GON

0.38

ênero

0.38

icoli

0.38

 Audiodateien

0.38

 consciência

0.37

("}");

0.37

 dovrebbero

0.37

Activations Density 0.073%

references to technical details and specifications related to components and processes

need for subsequent additional

No Comments

No Known Activations