INDEX

Explanations

references to specific terms or identifiers related to products and processes

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

technical terms and names

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

EDEFAULT

-0.44

 Taktlose

-0.43

 doInBackground

-0.38

AfterClass

-0.37

makeConstraints

-0.37

AddHtmlAttribute

-0.36

 constexpr

-0.36

]")]

-0.34

WriteTagHelper

-0.34

acamata

-0.34

POSITIVE LOGITS

Rüyada

0.54

ViewFeatures

0.49

MigrationBuilder

0.47

 ISTAT

0.46

Chemical

0.45

 Apostles

0.45

 Engines

0.45

 Predators

0.44

 apostles

0.44

Ses

0.44

Activations Density 0.278%

references to specific terms or identifiers related to products and processes

technical terms and names

No Comments

No Known Activations