unparallelled commited on
Commit
83fe46a
·
verified ·
1 Parent(s): a242065

Upload docs/TEARDOWN_READY_l3long_2026-09-09.txt with huggingface_hub

Browse files
docs/TEARDOWN_READY_l3long_2026-09-09.txt CHANGED
@@ -18,6 +18,20 @@ UPLOADED
18
  checkpoints/l3long_final_run17_step_2450_rank{0,1}.ckpt
19
  checkpoints/l3long_final_run17_step_2450_rank{0,1}.zero_g1 <- optimizer
20
  state, so step 2450 is RESUMABLE
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21
 
22
  TO RESUME l3long LATER
23
  Pull the four step_2450 files, restore them as
@@ -30,10 +44,8 @@ STILL NOT DONE
30
  2. ext_32k.sh (parent l2, b3 data) -- written, never run. NOTE a DIFFERENT
31
  32k extension (ext32k_b4, b4 data) WAS run and evaluated earlier today;
32
  see docs/EXT32K_RESULTS_2026-09-09.md and logs/ext32k/.
33
- 3. bluscript_amp1.cpp and the dist/Context_Parallelism submodule were left
34
- MODIFIED and UNCOMMITTED. I cannot run git per project rules. The source
35
- tarballs above are the only backup -- commit them yourself before relying
36
- on git history.
37
 
38
  NOT UPLOADED (deliberate, re-derivable)
39
  lendata/ 3.3 GB regenerate with fetch_eval_assets.sh
 
18
  checkpoints/l3long_final_run17_step_2450_rank{0,1}.ckpt
19
  checkpoints/l3long_final_run17_step_2450_rank{0,1}.zero_g1 <- optimizer
20
  state, so step 2450 is RESUMABLE
21
+ checkpoints/l3long1000_run17_step_1000_rank{0,1}.zero_g1
22
+ checkpoints/l3long2000_run17_step_2000_rank{0,1}.zero_g1
23
+ checkpoints/l3_blumodelcp_run1_step_400_rank{0,1}.{ckpt,zero_g1}
24
+ ^ l3, the eval baseline. It was the ONE genuine gap found in the final
25
+ audit: HF only had lenbucket_l1_blumodelcp_run1_step_400.ckpt (5.23
26
+ GB), a DIFFERENT file from local blumodelcp_run1_step_400.ckpt (8.18
27
+ GB). l3 appears in every eval table, so it is now backed up.
28
+ logs/misc_local_artifacts_2026-09-09.tgz 5.4 MB: b3_logs, setup_logs,
29
+ mspoe_2026-08-21, mspoe_run100_native, probe_run4,
30
+ context_rot_2026-08-19, CP_BluScreipt_Training_logs, Tests, Scripts,
31
+ context_parallel, graph.ipynb, teardown_logs. These were the local
32
+ dirs with no HF copy. decide_len_{4096,8192,16360} were already on HF
33
+ (26 files each) -- that is the MATCHED decision eval whose l2 numbers
34
+ the 32k results doc cites.
35
 
36
  TO RESUME l3long LATER
37
  Pull the four step_2450 files, restore them as
 
44
  2. ext_32k.sh (parent l2, b3 data) -- written, never run. NOTE a DIFFERENT
45
  32k extension (ext32k_b4, b4 data) WAS run and evaluated earlier today;
46
  see docs/EXT32K_RESULTS_2026-09-09.md and logs/ext32k/.
47
+ 3. Source changes: user confirmed pushed to git 2026-09-09. Source tarballs
48
+ on HF remain as a second copy.
 
 
49
 
50
  NOT UPLOADED (deliberate, re-derivable)
51
  lendata/ 3.3 GB regenerate with fetch_eval_assets.sh