← All tools

Backprop Trace

Org-only experimental cli TypeScript

Deterministic 26-rule verifier for neural-network training steps.

Deterministic 26-rule verifier for neural-network training steps. Re-derives gradients + optimizer state from named factors; emits canonical JSONL. SGD/Adam/AdamW/SGD-momentum (+coupled-L2 weight decay); live PyTorch + JAX helpers; CPU-only.

Backprop Trace screenshot preview
View on GitHub Issues Releases
Language TypeScript
Last updated Sep 7, 2026
Not in registry Install cmd No releases Documented
backpropagationcanonical-emissionclideterministicjsonlmazur

Related tools

Bytefit Hardware-aware local-LLM loadout planner: chooses model class, quant, KV-cache, context, and offload policy for your VRAM/RAM, and refuses configs that would silently page to disk. Brand Centralized brand asset management — migration, audit, org reconciliation, and integrity… Comfy Preflight A gate that runs on a ComfyUI workflow graph in the seconds before submission and halts a submission that would spend credits producing a known-wrong result.