Upload README.md with huggingface_hub

README.md

@@ -24,16 +24,22 @@ walkthrough notebook's `write_state` and consumed by its `load_state`:
 - `config` — the full `StackConfig` (asserted on load)
 - `code` — the notebook source that produced the state
 
+Sign convention (since 2026-09-01): the backward pass is seeded with
+`logits_grad = (onehot - probs) * softcap' * loss_scale`, so `grad`/`first_mntm`
+hold **update directions** (the negated loss gradient) and both optimizers apply
+`w + update`. States saved before this date carry `first_mntm` under the old sign
+and should not be loaded into the current notebook.
+
 Produced by the `DecoderStack d12 Walkthrough` notebook: Part 1 trains under the
 real 1680-step schedule, stops at `cfg.walkthrough_step`, saves + pushes here; the
 walkthrough part reloads and spells the last layer's forward/backward and one Muon
 step out flat.
 
-## Current states (run `40GB-A100_d12_walkthrough`, 2026-08-31)
+## Current states (run `40GB-A100_d12_walkthrough`, 2026-09-01)
 
 | file | where in the run | val bpb |
 |---|---|---|
-| `state_step000250.pt` | step 250 of 1680 (past all warmups, lr at peak) | 1.076657 |
+| `state_step000250.pt` | step 250 of 1680 (past all warmups, lr at peak) | 1.077130 |
 | `state_step000015.pt` | step 15 (mid lr/scalar/lm-head warmup) | 1.905353 |
 
 To walk through a different state, set `cfg.walkthrough_step` to its step number