numerics-debugging

name	numerics_debugging
description	Capture and compare per-op activations between two TorchTitan runs to spot numerics divergence (eager vs aot_fx_trace, FSDP vs no-FSDP, before vs after a refactor). Use when the user wants to debug bitwise / numeric drift in training, or invokes /numerics_debugging.

Numerics Debugging (DebugMode-based)

Per-op activation capture + comparison toolkit. Captures activations on a designated step via torch.utils._debug_mode.DebugMode, then diffs two captures into an HTML report to surface numerics divergence between runs that should agree (bitwise, or within float32 reduction-order noise).

Two pieces live in scripts/:

• activation_tracer.py — runtime capture, driven from Profiler by ActivationCaptureProfiler. Output: {dump_folder}/numerics/rank_{N}_activations.log.
• compare_numerics.py — diffs two logs, produces an HTML report. Standard library only, no torch import.

They sit outside the torchtitan package on purpose. Nothing in core torchtitan or graph_trainer references them; an agent must edit torchtitan to wire the tracer in before a capture run, and revert the edits when done (they don't belong on main). Because .claude is not a valid Python package name, activation_tracer is imported by putting scripts/ on sys.path rather than by dotted module path — see references/patching.md.

The two runs being compared must use the same dtype and seed. The matcher keys on shape + float64 L1 norm; a precision change (bf16 vs fp32) makes every row diverge and the matcher degrades to the structural-only stats pass.

Workflow

1. Patch torchtitan to wire the capture into Profiler (and graph_trainer if you're capturing the traced path). Full patch set: references/patching.md.
2. Capture twice, once per run you want to compare:

   ./run_train.sh \
       --dump_folder ./outputs/run_A \
       --training.steps 2 \
       --profiler.dump_numerics \
       --profiler.profile_freq 2 \
       --debug.seed 42 \
       --debug.deterministic \
       --training.mixed_precision_param float32
   

   The capture step is profile_freq. With profile_freq=2 and training.steps=2, step 1 warms up and step 2 is the snapshot. Capture adds ~10–40% memory only on the capture step (stats are computed inline in float64; tensors aren't held).
3. Diff the two logs:

   python .claude/skills/numerics_debugging/scripts/compare_numerics.py \
       outputs/run_A/numerics/rank_0_activations.log \
       outputs/run_B/numerics/rank_0_activations.log \
       --name1 run_A --name2 run_B \
       -o diff.html
   

   Open diff.html. Each row pairs one op from each run; cells turn red when a stat diverges; the "Match method" chip shows which of the four matching passes (override / exact key / fuzzy key / stats) paired the row.

Customizing

Excluded ops, numel / dtype filter, hash function, the manual-override file format, and HTML appearance are all tunable. See references/customization.md, which also catalogs the common eager-vs-traced mismatch patterns (AC-recompute FQN drift, per-layer counter shifts, collective renaming) you'll see in the diff and how to express them as overrides.