Skip to main content
Numerical testing ensures that EMEP tensor operations and merge strategies produce stable, predictable results across dtypes. Tolerance thresholds are defined per dtype and enforced in unit tests, merge tests, and integration tests.

Tolerance Thresholds

NaN and Inf Policy

EMEP tensor operations enforce a strict NaN/Inf policy:
  • NaN detection: any NaN in an output tensor triggers FAIL status.
  • Inf detection: finite Inf values are allowed only in intermediate softmax or attention scores if explicitly masked. Persistent Inf in weights or logits triggers FAIL.
  • Propagation: NaN must not propagate across layers. If a NaN is detected, the operation aborts and logs the layer and tensor name.

Numerical Stability Checks

The following checks run automatically during merge and evaluation:
  • Gradient clipping: if gradients exceed a threshold, they are clipped and logged.
  • Activation scaling: attention scores are scaled by sqrt(head_dim) to prevent overflow.
  • Weight initialization: LoRA matrices A and B are initialized with small variance to prevent explosion.

Cross-Reference to Math Specifications

Numerical stability is grounded in the math specifications:
  • SLERP Math: spherical interpolation stability near antipodal vectors.
  • TIES Math: sign consensus and trimming stability.
  • DARE Math: drop-rate scaling and rescaling correctness.
  • Task Arithmetic: linear combination stability.
  • Tensor Math: broadcasting, dtype coercion, and shape validation.

Test Harness

The numerical test harness:
  1. Generates random input tensors with fixed seeds.
  2. Runs the operation under test.
  3. Compares against a reference implementation (usually fp32 on CPU).
  4. Reports max absolute error, max relative error, and NaN/Inf counts.

Integration Points

  • Unit Test Plan: numerical tests are a required category for TensorEngine and MergeStrategy.
  • Model Merge Tests: merge test tolerances reference this specification.
  • CI: numerical tests run on every commit.
  • ArtifactStore: stores test matrices, reference outputs, and failure artifacts.