v41 quant worker: bootstrap-20260910T114946Z.log

runs/20260910T114946Z/bootstrap-20260910T114946Z.log

@@ -0,0 +1,115 @@
+=== v41 quant worker start 20260910T114946Z
+--- phase: setup (installing toolchain)
+WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv
+
+[notice] A new release of pip is available: 24.0 -> 26.2.1
+[notice] To update, run: pip install --upgrade pip
+--- phase: bundle (fetching worker bundle)
+--- phase: llama (cloning llama.cpp refs/pull/28696/head)
+cd62801 convert : add DeepSeek V4.1 (DeepseekV41ForCausalLM)
+--- phase: ggml (building ggml (CPU))
+ggml build rc=0
+--- phase: verify (verifying artifact digests)
+verify: OK
+--- phase: audit (fidelity audit)
+{
+ "IQ2_XXS": {
+  "tensors": 8,
+  "weights": 94371840,
+  "cosine_min": 0.94068,
+  "cosine_mean": 0.940953,
+  "relative_rms_error_max": 0.352575,
+  "relative_rms_error_mean": 0.35169
+ },
+ "Q2_K": {
+  "tensors": 8,
+  "weights": 94371840,
+  "cosine_min": 0.963304,
+  "cosine_mean": 0.963402,
+  "relative_rms_error_max": 0.270137,
+  "relative_rms_error_mean": 0.269774
+ }
+}
+pushed audit.json
+pushed verify.txt
+pushed llama-ref.txt
+--- phase: mapping (gguf mapping report)
+mapping rc=0
+{
+ "artifact_tensors": 49993,
+ "gguf_tensors": 1105,
+ "by_ggml_type": {
+  "BF16": 598,
+  "F32": 387,
+  "IQ2_XXS": 80,
+  "Q2_K": 40
+ },
+ "expert_tensors_aggregated": 120,
+ "skipped": {
+  "mmproj": 266,
+  "consumed by FP8 dequant": 1505,
+  "DSpark-only, not in deepseek4 tensor table": 5,
+  "DSpark/nextn expert aggregation unresolved": 1152
+ },
+ "unmapped": [],
+ "fp8_block": [
+  32,
+  32
+ ]
+}
+pushed gguf-pack-report.json
+--- phase: pack (writing GGUF)
+{
+ "artifact_tensors": 49993,
+ "gguf_tensors": 1105,
+ "by_ggml_type": {
+  "BF16": 598,
+  "F32": 387,
+  "IQ2_XXS": 80,
+  "Q2_K": 40
+ },
+ "expert_tensors_aggregated": 120,
+ "skipped": {
+  "mmproj": 266,
+  "consumed by FP8 dequant": 1505,
+  "DSpark-only, not in deepseek4 tensor table": 5,
+  "DSpark/nextn expert aggregation unresolved": 1152
+ },
+ "unmapped": [],
+ "fp8_block": [
+  32,
+  32
+ ]
+}
+DSV41-mixedq2-00001-of-00005.gguf: 245 tensors, 39.17 GB
+DSV41-mixedq2-00002-of-00005.gguf: 259 tensors, 39.99 GB
+DSV41-mixedq2-00003-of-00005.gguf: 235 tensors, 39.00 GB
+DSV41-mixedq2-00004-of-00005.gguf: 297 tensors, 39.58 GB
+DSV41-mixedq2-00005-of-00005.gguf: 69 tensors, 12.17 GB
+Traceback (most recent call last):
+  File "/workspace/v41-quant/scripts/v41_quant_to_gguf.py", line 786, in <module>
+    sys.exit(main())
+             ^^^^^^
+  File "/workspace/v41-quant/scripts/v41_quant_to_gguf.py", line 779, in main
+    validation = validate_gguf(args.out_dir)
+                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "/workspace/v41-quant/scripts/v41_quant_to_gguf.py", line 659, in validate_gguf
+    reader = GGUFReader(path)
+             ^^^^^^^^^^^^^^^^
+  File "/opt/llama.cpp/gguf-py/gguf/gguf_reader.py", line 194, in __init__
+    self._build_tensors(offs, tensors_fields)
+  File "/opt/llama.cpp/gguf-py/gguf/gguf_reader.py", line 343, in _build_tensors
+    raise ValueError(f'Found duplicated tensor with name {tensor_name}')
+ValueError: Found duplicated tensor with name output_norm
+worker aborted rc=1 at line 114
+--- phase: sums (hashing shards)
+5 SHA256SUMS
+{
+ "total_bytes": 169917778504,
+ "shards": 5
+}
+pushed pack.txt
+pushed gguf-summary.json
+pushed SHA256SUMS
+--- phase: 10 (pack rc=1 verify rc=0)
+--- phase: failed (pack rc=1)