Upload data3/is_sci_prompt1.txt with huggingface_hub

data3/is_sci_prompt1.txt

@@ -0,0 +1,30 @@
+You will receive the full contents of a code file.
+
+Your task is to decide whether this code is related to scientific computing.
+
+Decision criteria (be very strict):
+1. Only if the code clearly and explicitly contains academic concepts or terminology from natural sciences or scientific computing should it be considered scientific computing. Examples (including in comments, variable names, function names, strings):
+   - Chemistry-related: molar mass, stoichiometry, enthalpy, entropy, Gibbs free energy, pH, titration, reaction rate, bond length, molecule, atom, orbital, spectroscopy, chromatography, equilibrium constant, reaction kinetics, electrochemistry, etc.
+   - Physics-related: velocity, acceleration, force, momentum, kinetic energy, potential energy, Lagrangian, Hamiltonian, Schrödinger, quantum, spin, wavefunction, wavelength, frequency (in wave/EM context), thermodynamics, statistical mechanics, electromagnetism, Maxwell, fluid dynamics, Navier-Stokes, optics, diffraction, etc.
+   - Other clear scientific computing / numerical methods terms: simulation of physical/chemical systems, finite element, finite volume, molecular dynamics, DFT (density functional theory), Monte Carlo (for physical/statistical simulations), N-body simulation, PDE (partial differential equation), ODE solver, Runge-Kutta, discretization of physical laws, etc.
+2. These terms must appear in a clearly academic or scientific context, not as metaphors, business jargon, or generic words. Ambiguous words (e.g., energy, force, pressure, temperature, signal, spectrum) do NOT count unless they are obviously used in a natural science context (e.g., referencing known laws, equations, units, or phenomena).
+3. Pure engineering / software / infrastructure code is NOT scientific computing, even if it is used in scientific projects. If the content is about:
+   - business logic, web backends, APIs, databases, ETL, configuration, logging
+   - generic “data analysis”, “ML model”, “feature”, “score”, “metric” without explicit natural science formulas or laws
+   - robotics/control code without explicit physics formulas or academic terminology
+   then it MUST be treated as NOT scientific computing.
+4. Do NOT infer from:
+   - library names, package names, framework names
+   - file paths, project names, or repository names
+   - your external knowledge of what a library or tool is usually used for
+   Only judge based on explicit terms and formulas visible in the code itself.
+5. When in doubt (unclear, indirect, or weak evidence), always answer that it is NOT scientific computing.
+
+Output requirements:
+- If and only if the code IS scientific computing related under the strict criteria above, output exactly: true
+- Otherwise, output exactly: false
+- Do NOT output anything else: no explanations, no spaces, no quotes, no punctuation.
+
+Now analyze the following code file content:
+
+{CODE_FILE}