Add files using upload-large-folder tool

README.md

@@ -40,7 +40,7 @@ m1_mix_submit/
 ├── README.md                                   # this file
 ├── task_instructions.json                      # verbatim --global_task per task + scores
 ├── checkpoint/
-│   ├── m1_mix_final_step50000.pt.part00 … part03   # 15.3 GB full training ckpt, split into 4×≤4 GB parts
+│   ├── m1_mix_final_step50000.pt.part00 … part08   # 15.3 GB full training ckpt, split into 9 parts (2×4 GB + 7×≤1 GB)
 │   ├── m1_mix_final_step50000.pt.sha256        # SHA-256 of the reassembled checkpoint
 │   └── README_REASSEMBLE.md                    # how to cat the parts back together + verify
 ├── norm_stats/
@@ -60,9 +60,9 @@ m1_mix_submit/
 
 ### About the checkpoint
 
-> **Reassemble first.** The 15.3 GB checkpoint is uploaded as 4 byte-split parts
-> (`m1_mix_final_step50000.pt.part00…03`) because the upload path capped single files at
-> ~9.7 GB. Concatenation reproduces the original **bit-for-bit**:
+> **Reassemble first.** The 15.3 GB checkpoint is uploaded as 9 byte-split parts
+> (`m1_mix_final_step50000.pt.part00…08`) because the upload path capped single files and
+> throttled per-window bytes. Concatenation reproduces the original **bit-for-bit**:
 >
 > ```bash
 > cat m1_mix_final_step50000.pt.part?? > m1_mix_final_step50000.pt

checkpoint/README_REASSEMBLE.md

@@ -1,16 +1,23 @@
 # Reassembling the checkpoint
 
-The full training checkpoint `m1_mix_final_step50000.pt` (15.3 GB) is uploaded as **4
-byte-split parts** because the upload path capped single files at ~9.7 GB. The parts are a
-plain `split -b 4G` of the original file, so concatenating them in order reproduces the
-checkpoint **bit-for-bit**.
-
-| part                                    | size      |
-|-----------------------------------------|-----------|
-| `m1_mix_final_step50000.pt.part00`      | 4.0 GiB   |
-| `m1_mix_final_step50000.pt.part01`      | 4.0 GiB   |
-| `m1_mix_final_step50000.pt.part02`      | 4.0 GiB   |
-| `m1_mix_final_step50000.pt.part03`      | ~2.25 GiB |
+The full training checkpoint `m1_mix_final_step50000.pt` (15.3 GB) is uploaded as **9
+byte-split parts**. The upload path (an egress gateway) both capped single files at
+~9.7 GB *and* throttled cumulative bytes per connection window, so the file is split into
+small parts that each fit a window: the **first two parts are 4 GiB** (uploaded before the
+throttle was understood) and the **remaining seven are ≤1 GiB**. Concatenating all parts
+in sorted order reproduces the checkpoint **bit-for-bit**.
+
+| part                                    | size       |
+|-----------------------------------------|------------|
+| `m1_mix_final_step50000.pt.part00`      | 4.0 GiB    |
+| `m1_mix_final_step50000.pt.part01`      | 4.0 GiB    |
+| `m1_mix_final_step50000.pt.part02`      | 1.0 GiB    |
+| `m1_mix_final_step50000.pt.part03`      | 1.0 GiB    |
+| `m1_mix_final_step50000.pt.part04`      | 1.0 GiB    |
+| `m1_mix_final_step50000.pt.part05`      | 1.0 GiB    |
+| `m1_mix_final_step50000.pt.part06`      | 1.0 GiB    |
+| `m1_mix_final_step50000.pt.part07`      | 1.0 GiB    |
+| `m1_mix_final_step50000.pt.part08`      | ~0.25 GiB  |
 
 ## 1. Reassemble
 
@@ -18,7 +25,8 @@ checkpoint **bit-for-bit**.
 cat m1_mix_final_step50000.pt.part?? > m1_mix_final_step50000.pt
 ```
 
-(`part??` expands to `part00 part01 part02 part03` in sorted order.)
+(`part??` expands to `part00 part01 … part08` in sorted order — the command is unchanged
+regardless of how many parts there are.)
 
 ## 2. Verify integrity