rrm workspace: TRM/HRM/SRM code, Maze dataset, dynamical-analysis pipelineHEADmain

Curated export for clone-and-run Maze training (2x A6000) + diagnostics.
trm/hrm pretrain.py carry trajectory-augmentation code (backward-compatible).
Heavy artifacts (checkpoints/wandb/npz) gitignored; see PROVENANCE.md.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

1 files changed, 2 insertions, 0 deletions

diff --git a/research/flossing/ptrm_lambda_main_trm_base260410_k100_d64_sigma03_Lonly_n256.summary.csv b/research/flossing/ptrm_lambda_main_trm_base260410_k100_d64_sigma03_Lonly_n256.summary.csv
new file mode 100644
index 0000000..da911d6
--- /dev/null
+++ b/research/flossing/ptrm_lambda_main_trm_base260410_k100_d64_sigma03_Lonly_n256.summary.csv
@@ -0,0 +1,2 @@
+deterministic/exact,deterministic/token_acc,fd_lyap,include_clean,lambda_fail_mean,lambda_mean,lambda_success_mean,lyap_min/exact,lyap_min/token_acc,mean_rollout/exact,mean_rollout/token_acc,n_samples,noise_std,oracle_pass/exact,oracle_pass/token_acc,q_fail_mean,q_max/exact,q_max/token_acc,q_mean,q_minus_0.25lambda/exact,q_minus_0.25lambda/token_acc,q_minus_0.5lambda/exact,q_minus_0.5lambda/token_acc,q_minus_1lambda/exact,q_minus_1lambda/token_acc,q_minus_2lambda/exact,q_minus_2lambda/token_acc,q_success_mean,rollout0/exact,rollout0/token_acc,rollouts

+0.80078125,0.9281443357467651,1.0,0.0,7.016035556793213,5.942806243896484,5.796727180480957,0.89453125,0.9583333134651184,0.8801953196525574,0.9530397057533264,256.0,0.3,0.95703125,0.9853395223617554,-9.493144989013672,0.953125,0.9805652499198914,5.361981391906738,0.921875,0.9693769812583923,0.921875,0.9696663022041321,0.90625,0.9643614888191223,0.90234375,0.9642168283462524,7.383934020996094,0.8828125,0.9552469253540039,100.0