evalstate HF Staff commited on
Commit
4b4c7c4
·
verified ·
1 Parent(s): 231975f

Publish analysis hybrid-model-20260622t000017z for snapshot 20260622T000017Z

Browse files
analysis/current/analysis-report-hybrid.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "schema_version": "1.0",
3
  "repo": "huggingface/transformers",
4
- "snapshot_id": "20260621T180014Z",
5
- "generated_at": "2026-06-21T18:06:00Z",
6
  "evidence_quality": "full",
7
  "llm_enrichment": true,
8
  "meta_bugs": [
 
1
  {
2
  "schema_version": "1.0",
3
  "repo": "huggingface/transformers",
4
+ "snapshot_id": "20260622T000017Z",
5
+ "generated_at": "2026-06-22T00:04:29Z",
6
  "evidence_quality": "full",
7
  "llm_enrichment": true,
8
  "meta_bugs": [
analysis/current/analysis-report-hybrid.llm-reviews.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "schema_version": "1.0",
3
  "repo": "huggingface/transformers",
4
- "snapshot_id": "20260621T180014Z",
5
- "generated_at": "2026-06-21T18:06:00Z",
6
  "model": "gpt-5.4-mini?service_tier=flex",
7
  "reviews": [
8
  {
@@ -5067,8 +5067,8 @@
5067
  "review_unit_count": 58,
5068
  "status": "reviewed",
5069
  "reason": null,
5070
- "source": "cache",
5071
- "cache_hit": true,
5072
  "model": "gpt-5.4-mini?service_tier=flex",
5073
  "cluster_id": "cluster-10000-484",
5074
  "nodes": [
@@ -5152,7 +5152,7 @@
5152
  "issue:45081|issue:45356",
5153
  "issue:45356|issue:45907"
5154
  ],
5155
- "prepared_review_unit_hash": "3486db22b1d9289622ca0b018076784134b03457d12be58a7cd8a8fdc9a83041",
5156
  "component_budget": {
5157
  "node_count": 484,
5158
  "item_count": 484,
@@ -5174,189 +5174,195 @@
5174
  "aggressively_trimmed": false,
5175
  "split": true,
5176
  "analyst_result": {
5177
- "summary": "Most items are unrelated single-bug reports across different models or subsystems. Only two soft pairs look like true duplicates: the HF_HUB_OFFLINE tokenizer/cache regression (42886/44843) and the exact Qwen2_5_VLProcessor batched padding crash (44514/44545).",
5178
- "confidence": 0.79,
5179
- "canonical_issue_reason": "issue:44843 is the best canonical issue because it names the concrete failing code path (_patch_mistral_regex calling model_info() unconditionally) and clearly explains the HF_HUB_OFFLINE breakage.",
5180
  "canonical_pr_reason": null,
5181
- "best_issue_reason": "issue:44843 is the strongest overall issue candidate: clear root cause, clear user-facing failure, and the accepted duplicate pair points to the same offline tokenizer bug.",
5182
  "best_pr_reason": null,
5183
  "soft_edge_verdicts": [
5184
  {
5185
  "left": "issue:44625",
5186
  "right": "issue:45479",
5187
  "accept": false,
5188
- "reason": "Different bugs: one is config propagation for num_labels, the other is a zero-loss classification pathology."
5189
  },
5190
  {
5191
  "left": "issue:43644",
5192
  "right": "issue:46665",
5193
  "accept": false,
5194
- "reason": "Both mention loading/model behavior, but one is non-persistent buffers and the other is from_pretrained behavior in classes; not the same bug."
5195
  },
5196
  {
5197
  "left": "issue:44898",
5198
  "right": "issue:45072",
5199
  "accept": false,
5200
- "reason": "Different failure modes and code paths: Perceiver image-size handling vs bfloat16 dtype mismatch in other models."
5201
  },
5202
  {
5203
  "left": "issue:44062",
5204
  "right": "issue:44451",
5205
  "accept": false,
5206
- "reason": "The first is a tokenizer AddedToken TypeError; the second is a model-loading failure for a specific checkpoint."
5207
  },
5208
  {
5209
  "left": "issue:44987",
5210
  "right": "issue:45042",
5211
  "accept": false,
5212
- "reason": "One is a model-loading regression for a specific repo, the other is an image-processor dependency regression."
5213
  },
5214
  {
5215
  "left": "issue:43381",
5216
  "right": "issue:43844",
5217
  "accept": false,
5218
- "reason": "Eval-mode gradient checkpointing limitation is unrelated to the ZeRO-3 gradient blow-up issue."
5219
  },
5220
  {
5221
  "left": "issue:44368",
5222
  "right": "issue:45357",
5223
  "accept": false,
5224
- "reason": "A warning about tie_word_embeddings is unrelated to incorrect visual encoder keys on save_pretrained."
5225
  },
5226
  {
5227
  "left": "issue:39692",
5228
  "right": "issue:44610",
5229
  "accept": false,
5230
- "reason": "Doc example/model mismatch for SigLIP2 is different from OmDet-Turbo producing the wrong image resolution."
5231
  },
5232
  {
5233
  "left": "issue:43531",
5234
  "right": "issue:43645",
5235
  "accept": false,
5236
- "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to notebook custom-model initialization in v5.0.0."
5237
  },
5238
  {
5239
  "left": "issue:41720",
5240
  "right": "issue:45704",
5241
  "accept": false,
5242
- "reason": "CUDA assert with Qwen3 auto device mapping is not the same bug as T5 apex RMSNorm memory leak."
5243
  },
5244
  {
5245
  "left": "issue:46655",
5246
  "right": "issue:46693",
5247
  "accept": false,
5248
- "reason": "Video-processor padding calculation and FA2 long-context performance regression are different issues."
5249
  },
5250
  {
5251
  "left": "issue:43746",
5252
  "right": "issue:46050",
5253
  "accept": false,
5254
- "reason": "PEFT checkpoint loading for GraniteSpeech is unrelated to sinq quantization save/load breakage."
5255
  },
5256
  {
5257
  "left": "issue:45704",
5258
  "right": "issue:45923",
5259
  "accept": false,
5260
- "reason": "A memory leak in T5 apex usage is unrelated to a missing supports_gradient_checkpointing flag."
5261
  },
5262
  {
5263
  "left": "issue:44918",
5264
  "right": "issue:44936",
5265
  "accept": false,
5266
- "reason": "TRL unpacking failure and trainer.evaluate() after train() are distinct trainer failures."
5267
  },
5268
  {
5269
  "left": "issue:43421",
5270
  "right": "issue:43927",
5271
  "accept": false,
5272
- "reason": "Runtime special-token post-processing and DiaConfig save/load corruption are different tokenizer/config bugs."
5273
  },
5274
  {
5275
  "left": "issue:43381",
5276
  "right": "issue:45704",
5277
  "accept": false,
5278
- "reason": "Eval-mode checkpointing restriction is not the same as the T5 apex memory leak."
5279
  },
5280
  {
5281
  "left": "issue:42886",
5282
  "right": "issue:44843",
5283
- "accept": true,
5284
- "reason": "Same underlying offline-tokenizer bug: both describe HF_HUB_OFFLINE failures caused by an unconditional model_info() path in tokenizer loading."
5285
  },
5286
  {
5287
  "left": "issue:45759",
5288
  "right": "issue:45923",
5289
  "accept": false,
5290
- "reason": "Composite Qwen config handling and a missing gradient-checkpointing flag are unrelated."
5291
  },
5292
  {
5293
  "left": "issue:44514",
5294
  "right": "issue:44545",
5295
  "accept": true,
5296
- "reason": "Exact same Qwen2_5_VLProcessor.apply_chat_template batched padding=False crash; these are duplicates of the same bug report."
5297
  },
5298
  {
5299
  "left": "issue:46032",
5300
  "right": "issue:46463",
5301
  "accept": false,
5302
- "reason": "Mamba2Mixer use_cache correctness is unrelated to T5 decoder-input error messaging."
5303
  },
5304
  {
5305
  "left": "issue:39692",
5306
  "right": "issue:43873",
5307
  "accept": false,
5308
- "reason": "SigLIP2 documentation/model mismatch is unrelated to quantization offloading behavior."
5309
  },
5310
  {
5311
  "left": "issue:43746",
5312
  "right": "issue:43873",
5313
  "accept": false,
5314
- "reason": "Local PEFT checkpoint loading and quantization offloading are different subsystems and failures."
5315
  },
5316
  {
5317
  "left": "issue:43994",
5318
  "right": "issue:44671",
5319
  "accept": false,
5320
- "reason": "Different model families and symptoms: SigLIP2 nonsensical outputs vs CamemBERT masked LM predictions in v5."
5321
  },
5322
  {
5323
  "left": "issue:43421",
5324
  "right": "issue:43653",
5325
  "accept": false,
5326
- "reason": "Special-token post-processor updates and BigBirdTokenizer mask-token registration are not the same defect."
5327
  },
5328
  {
5329
  "left": "issue:30064",
5330
  "right": "issue:36010",
5331
  "accept": false,
5332
- "reason": "Image processor segmentation-map handling is unrelated to the GenerationMixin import error."
5333
  },
5334
  {
5335
  "left": "issue:43782",
5336
  "right": "issue:45381",
5337
  "accept": false,
5338
- "reason": "Qwen3VL weight_only loading failure and Qwen2.5-VL vision_position_ids bug are different issues."
 
 
 
 
 
 
5339
  },
5340
  {
5341
  "left": "issue:45081",
5342
  "right": "issue:45356",
5343
  "accept": false,
5344
- "reason": "Both mention fix_mistral_regex, but one crashes on Mistral tokenizer loading while the other is a Kimi-K2.5 codec/regression report."
5345
  },
5346
  {
5347
  "left": "issue:45356",
5348
  "right": "issue:45907",
5349
  "accept": false,
5350
- "reason": "Tokenizer codec regression is unrelated to list_repo_files not forwarding revision."
5351
  }
5352
  ]
5353
  },
5354
  "evaluator_result": {
5355
- "accept": true,
5356
- "feedback": "The summary is well grounded in the packet, and the soft-edge calls are conservative overall. The two accepted duplicate pairs are supported by the titles and align with the same underlying bug classes (offline tokenizer loading and the Qwen2_5_VLProcessor batched padding crash). The remaining pairs are appropriately rejected as different subsystems or failure modes."
5357
  },
5358
  "evaluator_used": true,
5359
- "retried": false,
5360
  "accepted_nontrivial_soft_edge": true,
5361
  "error_kind": null,
5362
  "error_message": null
@@ -5673,8 +5679,8 @@
5673
  "review_unit_count": 58,
5674
  "status": "reviewed",
5675
  "reason": null,
5676
- "source": "cache",
5677
- "cache_hit": true,
5678
  "model": "gpt-5.4-mini?service_tier=flex",
5679
  "cluster_id": "cluster-10000-484",
5680
  "nodes": [
@@ -5755,7 +5761,7 @@
5755
  "issue:45987|issue:46192",
5756
  "issue:44671|issue:45701"
5757
  ],
5758
- "prepared_review_unit_hash": "0d23d704e96e1d43a9a0a7256de1bace2a6fcdf792408638049e340ca7f2c307",
5759
  "component_budget": {
5760
  "node_count": 484,
5761
  "item_count": 484,
@@ -5777,180 +5783,192 @@
5777
  "aggressively_trimmed": false,
5778
  "split": true,
5779
  "analyst_result": {
5780
- "summary": "Most items are unrelated single issues across tokenizer, model-loading, dtype, and video/audio subsystems. The only likely duplicate pair is the tokenizer spacing bug vs. the broader slow/fast tokenizer inconsistency report.",
5781
- "confidence": 0.84,
5782
- "canonical_issue_reason": "issue:28218 is the clearest concrete symptom report for the tokenizer added-token spacing bug; issue:29159 reads as a broader description of the same tokenizer inconsistency.",
5783
  "canonical_pr_reason": null,
5784
- "best_issue_reason": "issue:28218 is the best representative because it is specific, actionable, and likely the more concrete anchor for the duplicate pair.",
5785
  "best_pr_reason": null,
5786
  "soft_edge_verdicts": [
5787
  {
5788
  "left": "issue:34634",
5789
  "right": "issue:36010",
5790
  "accept": false,
5791
- "reason": "Different problems: BarkProcessor voice_preset vs GenerationMixin import failure."
5792
  },
5793
  {
5794
  "left": "issue:43606",
5795
  "right": "issue:45072",
5796
  "accept": false,
5797
- "reason": "Different subsystems and failures: CPU offload device mismatch vs bfloat16 dtype mismatch."
5798
  },
5799
  {
5800
  "left": "issue:45183",
5801
  "right": "issue:45479",
5802
  "accept": false,
5803
- "reason": "Validation error messaging vs sequence-classification loss behavior; not the same bug."
5804
  },
5805
  {
5806
  "left": "issue:45584",
5807
  "right": "issue:45834",
5808
  "accept": false,
5809
- "reason": "Whisper empty transcription failure and Kosmos2.5 OCR index error are unrelated."
5810
  },
5811
  {
5812
  "left": "issue:43716",
5813
  "right": "issue:44008",
5814
  "accept": false,
5815
- "reason": "Different model families and root causes: image preprocessor dtype mismatch vs variable name collision in Gemma3n forward."
5816
  },
5817
  {
5818
  "left": "issue:43950",
5819
  "right": "issue:45987",
5820
  "accept": false,
5821
- "reason": "from_pretrained buffer corruption and StaticCache seq_length return type are unrelated."
5822
  },
5823
  {
5824
  "left": "issue:44479",
5825
  "right": "issue:45834",
5826
  "accept": false,
5827
- "reason": "Different video regression in Qwen models vs Kosmos2.5 long-OCR index error."
5828
  },
5829
  {
5830
  "left": "issue:43582",
5831
  "right": "issue:46512",
5832
  "accept": false,
5833
- "reason": "Apple Silicon allocator TypeError vs stale config.dtype after from_config; different issues."
5834
  },
5835
  {
5836
  "left": "issue:43388",
5837
  "right": "issue:46032",
5838
  "accept": false,
5839
- "reason": "Metric gathering label truncation and Mamba2 cache correctness are unrelated."
5840
  },
5841
  {
5842
  "left": "issue:44991",
5843
  "right": "issue:46291",
5844
  "accept": false,
5845
- "reason": "Tokenizer loading failure for a specific model vs importing pipelines needing kernels version/revision."
5846
  },
5847
  {
5848
  "left": "issue:43577",
5849
  "right": "issue:43818",
5850
  "accept": false,
5851
- "reason": "BLIP2 dtype loading issue and Video-LLaVA architecture issue are unrelated."
5852
  },
5853
  {
5854
  "left": "issue:43825",
5855
  "right": "issue:43994",
5856
  "accept": false,
5857
- "reason": "Translation-support error message vs SigLIP2 inference corruption; same area but not same bug."
5858
  },
5859
  {
5860
  "left": "issue:46294",
5861
  "right": "issue:46463",
5862
  "accept": false,
5863
- "reason": "Qwen video-token mismatch and T5 decoder validation error are different concrete failures."
5864
  },
5865
  {
5866
  "left": "issue:43782",
5867
  "right": "issue:44977",
5868
  "accept": false,
5869
- "reason": "Qwen3VL weight_only loading error vs Qwen3.5 flash-attention generation issue."
5870
  },
5871
  {
5872
  "left": "issue:44561",
5873
  "right": "issue:45020",
5874
  "accept": false,
5875
- "reason": "Both mention remote_code compatibility, but they describe different API breakages and are too broad to merge."
5876
  },
5877
  {
5878
  "left": "issue:43421",
5879
  "right": "issue:44811",
5880
  "accept": false,
5881
- "reason": "Runtime special-token post-processor update bug vs Whisper batch_decode skip_special_tokens bug."
5882
  },
5883
  {
5884
  "left": "issue:44805",
5885
  "right": "issue:44936",
5886
  "accept": false,
5887
- "reason": "Mask shape IndexError vs trainer.evaluate() failure after train(); no clear duplicate relation."
5888
  },
5889
  {
5890
  "left": "issue:44610",
5891
  "right": "issue:45072",
5892
  "accept": false,
5893
- "reason": "OmDet-Turbo input-size mismatch and dtype mismatch in other models are unrelated."
5894
  },
5895
  {
5896
  "left": "issue:43720",
5897
  "right": "issue:45072",
5898
  "accept": false,
5899
- "reason": "Packed-weights unpacking failure in accelerate loading vs dtype mismatch in inference."
5900
  },
5901
  {
5902
  "left": "issue:43540",
5903
  "right": "issue:45381",
5904
  "accept": false,
5905
- "reason": "Video input processing error in Qwen3OmniMoe vs Qwen2.5-VL vision_position_ids regression."
5906
  },
5907
  {
5908
  "left": "issue:43825",
5909
  "right": "issue:43827",
5910
  "accept": false,
5911
- "reason": "One is an error-message issue, the other is stale documentation; related topic but not the same bug."
5912
  },
5913
  {
5914
  "left": "issue:28218",
5915
  "right": "issue:29159",
5916
  "accept": true,
5917
- "reason": "Both describe the same tokenizer added-token behavior problem: extra spacing / inconsistent slow-vs-fast tokenizer handling."
5918
  },
5919
  {
5920
  "left": "issue:43749",
5921
  "right": "issue:44291",
5922
  "accept": false,
5923
- "reason": "Different model-loading failures: FSDP CPU RAM efficient loading vs unexpected _is_hf_initialized arg with init_empty_weights."
5924
  },
5925
  {
5926
  "left": "issue:43645",
5927
  "right": "issue:46665",
5928
  "accept": false,
5929
- "reason": "Custom models in Jupyter notebook initialization vs from_pretrained inside Python classes are different reproducibility issues."
5930
  },
5931
  {
5932
  "left": "issue:43577",
5933
  "right": "issue:43582",
5934
  "accept": false,
5935
- "reason": "BLIP2 dtype mismatch and Apple Silicon allocator TypeError are unrelated."
5936
  },
5937
  {
5938
  "left": "issue:45987",
5939
  "right": "issue:46192",
5940
  "accept": false,
5941
- "reason": "Cache length return-type bug and return_tensors being ignored are different API regressions."
5942
  },
5943
  {
5944
  "left": "issue:44671",
5945
  "right": "issue:45701",
5946
  "accept": false,
5947
- "reason": "CamemBERT masked LM regression and general tokenization changes are not the same underlying bug."
 
 
 
 
 
 
 
 
 
 
 
 
5948
  }
5949
  ]
5950
  },
5951
  "evaluator_result": {
5952
  "accept": true,
5953
- "feedback": "The summary is grounded in the packet and stays appropriately cautious. The accepted tokenizer pair is plausible from the titles, and the other soft-edge verdicts are conservative rather than overstated."
5954
  },
5955
  "evaluator_used": true,
5956
  "retried": false,
@@ -8087,8 +8105,8 @@
8087
  "review_unit_count": 58,
8088
  "status": "reviewed",
8089
  "reason": null,
8090
- "source": "llm",
8091
- "cache_hit": false,
8092
  "model": "gpt-5.4-mini?service_tier=flex",
8093
  "cluster_id": "cluster-10000-484",
8094
  "nodes": [
@@ -11097,8 +11115,8 @@
11097
  "review_unit_count": 58,
11098
  "status": "reviewed",
11099
  "reason": null,
11100
- "source": "llm",
11101
- "cache_hit": false,
11102
  "model": "gpt-5.4-mini?service_tier=flex",
11103
  "cluster_id": "cluster-10000-484",
11104
  "nodes": [
@@ -11687,8 +11705,8 @@
11687
  "review_unit_count": 58,
11688
  "status": "reviewed",
11689
  "reason": null,
11690
- "source": "llm",
11691
- "cache_hit": false,
11692
  "model": "gpt-5.4-mini?service_tier=flex",
11693
  "cluster_id": "cluster-10000-484",
11694
  "nodes": [
@@ -12302,8 +12320,8 @@
12302
  "review_unit_count": 58,
12303
  "status": "reviewed",
12304
  "reason": null,
12305
- "source": "llm",
12306
- "cache_hit": false,
12307
  "model": "gpt-5.4-mini?service_tier=flex",
12308
  "cluster_id": "cluster-10000-484",
12309
  "nodes": [
@@ -12607,8 +12625,8 @@
12607
  "review_unit_count": 58,
12608
  "status": "reviewed",
12609
  "reason": null,
12610
- "source": "llm",
12611
- "cache_hit": false,
12612
  "model": "gpt-5.4-mini?service_tier=flex",
12613
  "cluster_id": "cluster-10000-484",
12614
  "nodes": [
@@ -13224,8 +13242,8 @@
13224
  "review_unit_count": 58,
13225
  "status": "reviewed",
13226
  "reason": null,
13227
- "source": "llm",
13228
- "cache_hit": false,
13229
  "model": "gpt-5.4-mini?service_tier=flex",
13230
  "cluster_id": "cluster-10000-484",
13231
  "nodes": [
@@ -15058,8 +15076,8 @@
15058
  "review_unit_count": 58,
15059
  "status": "reviewed",
15060
  "reason": null,
15061
- "source": "llm",
15062
- "cache_hit": false,
15063
  "model": "gpt-5.4-mini?service_tier=flex",
15064
  "cluster_id": "cluster-10000-484",
15065
  "nodes": [
@@ -16021,8 +16039,8 @@
16021
  "review_unit_count": 58,
16022
  "status": "reviewed",
16023
  "reason": null,
16024
- "source": "llm",
16025
- "cache_hit": false,
16026
  "model": "gpt-5.4-mini?service_tier=flex",
16027
  "cluster_id": "cluster-10000-484",
16028
  "nodes": [
@@ -20312,8 +20330,8 @@
20312
  "review_unit_count": 1,
20313
  "status": "reviewed",
20314
  "reason": null,
20315
- "source": "llm",
20316
- "cache_hit": false,
20317
  "model": "gpt-5.4-mini?service_tier=flex",
20318
  "cluster_id": "cluster-40546-2",
20319
  "nodes": [
@@ -20375,8 +20393,8 @@
20375
  "review_unit_count": 1,
20376
  "status": "reviewed",
20377
  "reason": null,
20378
- "source": "llm",
20379
- "cache_hit": false,
20380
  "model": "gpt-5.4-mini?service_tier=flex",
20381
  "cluster_id": "cluster-40755-2",
20382
  "nodes": [
@@ -20501,8 +20519,8 @@
20501
  "review_unit_count": 1,
20502
  "status": "reviewed",
20503
  "reason": null,
20504
- "source": "llm",
20505
- "cache_hit": false,
20506
  "model": "gpt-5.4-mini?service_tier=flex",
20507
  "cluster_id": "cluster-41855-14",
20508
  "nodes": [
@@ -20705,8 +20723,8 @@
20705
  "review_unit_count": 1,
20706
  "status": "reviewed",
20707
  "reason": null,
20708
- "source": "llm",
20709
- "cache_hit": false,
20710
  "model": "gpt-5.4-mini?service_tier=flex",
20711
  "cluster_id": "cluster-42975-2",
20712
  "nodes": [
@@ -20768,8 +20786,8 @@
20768
  "review_unit_count": 1,
20769
  "status": "reviewed",
20770
  "reason": null,
20771
- "source": "llm",
20772
- "cache_hit": false,
20773
  "model": "gpt-5.4-mini?service_tier=flex",
20774
  "cluster_id": "cluster-42978-7",
20775
  "nodes": [
@@ -20913,8 +20931,8 @@
20913
  "review_unit_count": 1,
20914
  "status": "reviewed",
20915
  "reason": null,
20916
- "source": "llm",
20917
- "cache_hit": false,
20918
  "model": "gpt-5.4-mini?service_tier=flex",
20919
  "cluster_id": "cluster-43291-8",
20920
  "nodes": [
@@ -21073,8 +21091,8 @@
21073
  "review_unit_count": 1,
21074
  "status": "reviewed",
21075
  "reason": null,
21076
- "source": "llm",
21077
- "cache_hit": false,
21078
  "model": "gpt-5.4-mini?service_tier=flex",
21079
  "cluster_id": "cluster-43488-3",
21080
  "nodes": [
@@ -21151,8 +21169,8 @@
21151
  "review_unit_count": 1,
21152
  "status": "reviewed",
21153
  "reason": null,
21154
- "source": "llm",
21155
- "cache_hit": false,
21156
  "model": "gpt-5.4-mini?service_tier=flex",
21157
  "cluster_id": "cluster-43612-2",
21158
  "nodes": [
@@ -21214,8 +21232,8 @@
21214
  "review_unit_count": 1,
21215
  "status": "reviewed",
21216
  "reason": null,
21217
- "source": "llm",
21218
- "cache_hit": false,
21219
  "model": "gpt-5.4-mini?service_tier=flex",
21220
  "cluster_id": "cluster-43651-2",
21221
  "nodes": [
@@ -21277,8 +21295,8 @@
21277
  "review_unit_count": 1,
21278
  "status": "reviewed",
21279
  "reason": null,
21280
- "source": "llm",
21281
- "cache_hit": false,
21282
  "model": "gpt-5.4-mini?service_tier=flex",
21283
  "cluster_id": "cluster-43656-2",
21284
  "nodes": [
@@ -21406,8 +21424,8 @@
21406
  "review_unit_count": 2,
21407
  "status": "reviewed",
21408
  "reason": null,
21409
- "source": "llm",
21410
- "cache_hit": false,
21411
  "model": "gpt-5.4-mini?service_tier=flex",
21412
  "cluster_id": "cluster-43996-28",
21413
  "nodes": [
@@ -21892,8 +21910,8 @@
21892
  "review_unit_count": 2,
21893
  "status": "reviewed",
21894
  "reason": null,
21895
- "source": "llm",
21896
- "cache_hit": false,
21897
  "model": "gpt-5.4-mini?service_tier=flex",
21898
  "cluster_id": "cluster-43996-28",
21899
  "nodes": [
@@ -21996,8 +22014,8 @@
21996
  "review_unit_count": 1,
21997
  "status": "reviewed",
21998
  "reason": null,
21999
- "source": "llm",
22000
- "cache_hit": false,
22001
  "model": "gpt-5.4-mini?service_tier=flex",
22002
  "cluster_id": "cluster-43998-13",
22003
  "nodes": [
@@ -22337,8 +22355,8 @@
22337
  "review_unit_count": 1,
22338
  "status": "reviewed",
22339
  "reason": null,
22340
- "source": "llm",
22341
- "cache_hit": false,
22342
  "model": "gpt-5.4-mini?service_tier=flex",
22343
  "cluster_id": "cluster-44662-2",
22344
  "nodes": [
@@ -22400,8 +22418,8 @@
22400
  "review_unit_count": 1,
22401
  "status": "reviewed",
22402
  "reason": null,
22403
- "source": "llm",
22404
- "cache_hit": false,
22405
  "model": "gpt-5.4-mini?service_tier=flex",
22406
  "cluster_id": "cluster-44979-2",
22407
  "nodes": [
@@ -22463,8 +22481,8 @@
22463
  "review_unit_count": 1,
22464
  "status": "reviewed",
22465
  "reason": null,
22466
- "source": "llm",
22467
- "cache_hit": false,
22468
  "model": "gpt-5.4-mini?service_tier=flex",
22469
  "cluster_id": "cluster-45086-2",
22470
  "nodes": [
@@ -22526,8 +22544,8 @@
22526
  "review_unit_count": 1,
22527
  "status": "reviewed",
22528
  "reason": null,
22529
- "source": "llm",
22530
- "cache_hit": false,
22531
  "model": "gpt-5.4-mini?service_tier=flex",
22532
  "cluster_id": "cluster-45115-3",
22533
  "nodes": [
@@ -22597,8 +22615,8 @@
22597
  "review_unit_count": 1,
22598
  "status": "reviewed",
22599
  "reason": null,
22600
- "source": "llm",
22601
- "cache_hit": false,
22602
  "model": "gpt-5.4-mini?service_tier=flex",
22603
  "cluster_id": "cluster-45332-2",
22604
  "nodes": [
@@ -22660,8 +22678,8 @@
22660
  "review_unit_count": 1,
22661
  "status": "reviewed",
22662
  "reason": null,
22663
- "source": "llm",
22664
- "cache_hit": false,
22665
  "model": "gpt-5.4-mini?service_tier=flex",
22666
  "cluster_id": "cluster-45415-2",
22667
  "nodes": [
@@ -22734,7 +22752,7 @@
22734
  "soft_pairs": [
22735
  "pull_request:45513|pull_request:46084"
22736
  ],
22737
- "prepared_review_unit_hash": "9a04aa5246def84ab1a92d520ea9a84120af716337f00ab1bd0484d5911c38d8",
22738
  "component_budget": {
22739
  "node_count": 2,
22740
  "item_count": 2,
@@ -22756,18 +22774,18 @@
22756
  "aggressively_trimmed": false,
22757
  "split": false,
22758
  "analyst_result": {
22759
- "summary": "These are unrelated model-specific fixes: one for Qwen3.5/Qwen3 Next GDN linear attention cached multi-token forward, the other for Mamba2 use_cache with seq_len>1. They do not describe the same bug or mergeable change.",
22760
  "confidence": 0.98,
22761
  "canonical_issue_reason": null,
22762
- "canonical_pr_reason": "PR 46084 is the stronger standalone representative: it has an explicit issue target, a narrowly scoped cache bug, and a focused fix path.",
22763
  "best_issue_reason": null,
22764
- "best_pr_reason": "PR 46084 is the best PR to surface in the cluster because it is concrete and self-contained; PR 45513 is a different model family and bug.",
22765
  "soft_edge_verdicts": [
22766
  {
22767
  "left": "pull_request:45513",
22768
  "right": "pull_request:46084",
22769
  "accept": false,
22770
- "reason": "Different model families and different failure modes: Qwen3.5 GDN linear attention multi-token cached forward vs Mamba2 use_cache seq_len>1. Not the same underlying change."
22771
  }
22772
  ]
22773
  },
@@ -22786,8 +22804,8 @@
22786
  "review_unit_count": 1,
22787
  "status": "reviewed",
22788
  "reason": null,
22789
- "source": "llm",
22790
- "cache_hit": false,
22791
  "model": "gpt-5.4-mini?service_tier=flex",
22792
  "cluster_id": "cluster-45548-2",
22793
  "nodes": [
@@ -22849,8 +22867,8 @@
22849
  "review_unit_count": 1,
22850
  "status": "reviewed",
22851
  "reason": null,
22852
- "source": "llm",
22853
- "cache_hit": false,
22854
  "model": "gpt-5.4-mini?service_tier=flex",
22855
  "cluster_id": "cluster-45639-2",
22856
  "nodes": [
@@ -22912,8 +22930,8 @@
22912
  "review_unit_count": 1,
22913
  "status": "reviewed",
22914
  "reason": null,
22915
- "source": "llm",
22916
- "cache_hit": false,
22917
  "model": "gpt-5.4-mini?service_tier=flex",
22918
  "cluster_id": "cluster-45720-2",
22919
  "nodes": [
@@ -23041,8 +23059,8 @@
23041
  "review_unit_count": 1,
23042
  "status": "reviewed",
23043
  "reason": null,
23044
- "source": "llm",
23045
- "cache_hit": false,
23046
  "model": "gpt-5.4-mini?service_tier=flex",
23047
  "cluster_id": "cluster-45932-2",
23048
  "nodes": [
@@ -23104,8 +23122,8 @@
23104
  "review_unit_count": 1,
23105
  "status": "reviewed",
23106
  "reason": null,
23107
- "source": "llm",
23108
- "cache_hit": false,
23109
  "model": "gpt-5.4-mini?service_tier=flex",
23110
  "cluster_id": "cluster-46126-2",
23111
  "nodes": [
@@ -23167,8 +23185,8 @@
23167
  "review_unit_count": 1,
23168
  "status": "reviewed",
23169
  "reason": null,
23170
- "source": "llm",
23171
- "cache_hit": false,
23172
  "model": "gpt-5.4-mini?service_tier=flex",
23173
  "cluster_id": "cluster-46138-10",
23174
  "nodes": [
@@ -23434,8 +23452,8 @@
23434
  "review_unit_count": 1,
23435
  "status": "reviewed",
23436
  "reason": null,
23437
- "source": "llm",
23438
- "cache_hit": false,
23439
  "model": "gpt-5.4-mini?service_tier=flex",
23440
  "cluster_id": "cluster-46152-2",
23441
  "nodes": [
@@ -23497,8 +23515,8 @@
23497
  "review_unit_count": 1,
23498
  "status": "reviewed",
23499
  "reason": null,
23500
- "source": "llm",
23501
- "cache_hit": false,
23502
  "model": "gpt-5.4-mini?service_tier=flex",
23503
  "cluster_id": "cluster-46154-2",
23504
  "nodes": [
@@ -23689,8 +23707,8 @@
23689
  "review_unit_count": 1,
23690
  "status": "reviewed",
23691
  "reason": null,
23692
- "source": "llm",
23693
- "cache_hit": false,
23694
  "model": "gpt-5.4-mini?service_tier=flex",
23695
  "cluster_id": "cluster-46310-2",
23696
  "nodes": [
@@ -23752,8 +23770,8 @@
23752
  "review_unit_count": 1,
23753
  "status": "reviewed",
23754
  "reason": null,
23755
- "source": "llm",
23756
- "cache_hit": false,
23757
  "model": "gpt-5.4-mini?service_tier=flex",
23758
  "cluster_id": "cluster-46376-2",
23759
  "nodes": [
@@ -23815,8 +23833,8 @@
23815
  "review_unit_count": 1,
23816
  "status": "reviewed",
23817
  "reason": null,
23818
- "source": "llm",
23819
- "cache_hit": false,
23820
  "model": "gpt-5.4-mini?service_tier=flex",
23821
  "cluster_id": "cluster-46574-2",
23822
  "nodes": [
@@ -23878,8 +23896,8 @@
23878
  "review_unit_count": 1,
23879
  "status": "reviewed",
23880
  "reason": null,
23881
- "source": "llm",
23882
- "cache_hit": false,
23883
  "model": "gpt-5.4-mini?service_tier=flex",
23884
  "cluster_id": "cluster-46603-2",
23885
  "nodes": [
@@ -23958,7 +23976,7 @@
23958
  "pull_request:46748|pull_request:46781",
23959
  "pull_request:46748|pull_request:46773"
23960
  ],
23961
- "prepared_review_unit_hash": "b4c1b3d2af157df30c9c6f134c28ffd72aa691ad21ccac03c737783fe18b6aff",
23962
  "component_budget": {
23963
  "node_count": 4,
23964
  "item_count": 4,
@@ -23980,42 +23998,42 @@
23980
  "aggressively_trimmed": false,
23981
  "split": false,
23982
  "analyst_result": {
23983
- "summary": "These are four distinct pull requests around hybrid/linear-attention cache and masking behavior, but they address different bugs and code paths. None look like true duplicates that could be merged into one PR.",
23984
- "confidence": 0.95,
23985
  "canonical_issue_reason": null,
23986
- "canonical_pr_reason": "PR 46748 is the best cluster representative because it targets shared cache infrastructure in `cache_utils.py` and has the clearest broad applicability to hybrid/offloaded cache behavior, unlike the model-specific fixes in the other PRs.",
23987
  "best_issue_reason": null,
23988
- "best_pr_reason": "46748 is the strongest representative of the cluster theme: it fixes a concrete cache utility bug with general hybrid-model impact and includes a test, while the others are narrower model-specific fixes in unrelated forward/masking paths.",
23989
  "soft_edge_verdicts": [
23990
  {
23991
  "left": "pull_request:46741",
23992
  "right": "pull_request:46781",
23993
  "accept": false,
23994
- "reason": "Different bugs and code paths: Mamba2/Zamba2 speculative-decoding chunked-prefill handling vs QuantizedCache initialization for hybrid models. Same broad area, but not the same underlying fix."
23995
  },
23996
  {
23997
  "left": "pull_request:46741",
23998
  "right": "pull_request:46773",
23999
  "accept": false,
24000
- "reason": "One fixes Mamba2 mixer chunked-prefill/speculative decoding; the other fixes padding-mask application for batch size 1 in linear-attention models. No shared concrete bug."
24001
  },
24002
  {
24003
  "left": "pull_request:46741",
24004
  "right": "pull_request:46748",
24005
  "accept": false,
24006
- "reason": "Model forward-path fix in Nemotron-H/Zamba2 versus cache prefetch/device-mismatch logic in `cache_utils.py`. Related only at a high level."
24007
  },
24008
  {
24009
  "left": "pull_request:46748",
24010
- "right": "pull_request:46773",
24011
  "accept": false,
24012
- "reason": "Both involve hybrid/linear-attention models, but one is about offloaded cache device mismatch and the other about padding-mask skipping on batch size 1. Distinct issues."
24013
  },
24014
  {
24015
  "left": "pull_request:46748",
24016
- "right": "pull_request:46781",
24017
  "accept": false,
24018
- "reason": "Both touch `cache_utils.py`, but they fix different cache subsystems: prefetching offloaded layers versus QuantizedCache layer construction for hybrid models. Not the same concrete bug."
24019
  }
24020
  ]
24021
  },
@@ -24034,8 +24052,8 @@
24034
  "review_unit_count": 1,
24035
  "status": "reviewed",
24036
  "reason": null,
24037
- "source": "llm",
24038
- "cache_hit": false,
24039
  "model": "gpt-5.4-mini?service_tier=flex",
24040
  "cluster_id": "cluster-46753-6",
24041
  "nodes": [
@@ -24174,8 +24192,8 @@
24174
  "review_unit_count": 1,
24175
  "status": "reviewed",
24176
  "reason": null,
24177
- "source": "llm",
24178
- "cache_hit": false,
24179
  "model": "gpt-5.4-mini?service_tier=flex",
24180
  "cluster_id": "cluster-46757-2",
24181
  "nodes": [
 
1
  {
2
  "schema_version": "1.0",
3
  "repo": "huggingface/transformers",
4
+ "snapshot_id": "20260622T000017Z",
5
+ "generated_at": "2026-06-22T00:04:29Z",
6
  "model": "gpt-5.4-mini?service_tier=flex",
7
  "reviews": [
8
  {
 
5067
  "review_unit_count": 58,
5068
  "status": "reviewed",
5069
  "reason": null,
5070
+ "source": "llm",
5071
+ "cache_hit": false,
5072
  "model": "gpt-5.4-mini?service_tier=flex",
5073
  "cluster_id": "cluster-10000-484",
5074
  "nodes": [
 
5152
  "issue:45081|issue:45356",
5153
  "issue:45356|issue:45907"
5154
  ],
5155
+ "prepared_review_unit_hash": "59b54eaff8d806f94dfd3e3339e25bd5186c29ae828367cd65f6064021f8ac2d",
5156
  "component_budget": {
5157
  "node_count": 484,
5158
  "item_count": 484,
 
5174
  "aggressively_trimmed": false,
5175
  "split": true,
5176
  "analyst_result": {
5177
+ "summary": "Only one soft pair looks like a true duplicate: the Qwen2_5_VLProcessor batched-input crash with padding=False. The remaining pairs are mostly related-sounding but point to different concrete failures, code paths, or model families.",
5178
+ "confidence": 0.92,
5179
+ "canonical_issue_reason": "issue:44514 is the earlier, clearer report of the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash; issue:44545 is a near-verbatim later duplicate.",
5180
  "canonical_pr_reason": null,
5181
+ "best_issue_reason": "issue:44514 is the best standalone issue to keep as the canonical representative because it is the original report of the exact crash scenario; issue:44545 adds follow-up activity but not a distinct bug.",
5182
  "best_pr_reason": null,
5183
  "soft_edge_verdicts": [
5184
  {
5185
  "left": "issue:44625",
5186
  "right": "issue:45479",
5187
  "accept": false,
5188
+ "reason": "Both are classification-related, but one is Qwen3.5 config propagation and the other is a generic single-label zero-loss bug across multiple models; different failure modes and fixes."
5189
  },
5190
  {
5191
  "left": "issue:43644",
5192
  "right": "issue:46665",
5193
  "accept": false,
5194
+ "reason": "Non-persistent buffer junk on v5.0.0 is unrelated to from_pretrained behavior inside Python classes; no shared code path is evident."
5195
  },
5196
  {
5197
  "left": "issue:44898",
5198
  "right": "issue:45072",
5199
  "accept": false,
5200
+ "reason": "Image-size interpolation failure and bf16 dtype mismatch are separate runtime problems in different model paths."
5201
  },
5202
  {
5203
  "left": "issue:44062",
5204
  "right": "issue:44451",
5205
  "accept": false,
5206
+ "reason": "AddedToken constructor argument conflict is a tokenizer API bug, while loading vesteinn/ScandiBERT is a model-specific loading failure."
5207
  },
5208
  {
5209
  "left": "issue:44987",
5210
  "right": "issue:45042",
5211
  "accept": false,
5212
+ "reason": "A specific model load regression and a PIL image-processor torchvision requirement are different bugs, even though both are version-related."
5213
  },
5214
  {
5215
  "left": "issue:43381",
5216
  "right": "issue:43844",
5217
  "accept": false,
5218
+ "reason": "Gradient checkpointing in eval mode and ZeRO-3 gradient growth are different training semantics issues."
5219
  },
5220
  {
5221
  "left": "issue:44368",
5222
  "right": "issue:45357",
5223
  "accept": false,
5224
+ "reason": "A LoRA warning about tie_word_embeddings is not the same as incorrect visual encoder keys being saved in Qwen3.5."
5225
  },
5226
  {
5227
  "left": "issue:39692",
5228
  "right": "issue:44610",
5229
  "accept": false,
5230
+ "reason": "SigLIP2 doc example errors and OmDet-Turbo processor size mismatch affect different multimodal flows."
5231
  },
5232
  {
5233
  "left": "issue:43531",
5234
  "right": "issue:43645",
5235
  "accept": false,
5236
+ "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to notebook-defined custom model initialization in v5."
5237
  },
5238
  {
5239
  "left": "issue:41720",
5240
  "right": "issue:45704",
5241
  "accept": false,
5242
+ "reason": "cudaErrorAssert during Qwen3 device mapping and an apex RMSNorm memory leak are distinct bugs."
5243
  },
5244
  {
5245
  "left": "issue:46655",
5246
  "right": "issue:46693",
5247
  "accept": false,
5248
+ "reason": "Video padding calculation in a processor is unrelated to the FA2 long-context performance regression."
5249
  },
5250
  {
5251
  "left": "issue:43746",
5252
  "right": "issue:46050",
5253
  "accept": false,
5254
+ "reason": "PEFT local-checkpoint loading and sinq quantization save/load breakage are different persistence paths."
5255
  },
5256
  {
5257
  "left": "issue:45704",
5258
  "right": "issue:45923",
5259
  "accept": false,
5260
+ "reason": "A T5 RMSNorm leak and a missing supports_gradient_checkpointing flag on trust_remote_code model are different issues."
5261
  },
5262
  {
5263
  "left": "issue:44918",
5264
  "right": "issue:44936",
5265
  "accept": false,
5266
+ "reason": "TRL SFT embedding unpacking and trainer.evaluate() failing after train() are separate trainer bugs."
5267
  },
5268
  {
5269
  "left": "issue:43421",
5270
  "right": "issue:43927",
5271
  "accept": false,
5272
+ "reason": "Runtime post-processor updates for special tokens are not the same as losing custom token IDs on save/load."
5273
  },
5274
  {
5275
  "left": "issue:43381",
5276
  "right": "issue:45704",
5277
  "accept": false,
5278
+ "reason": "Eval-mode checkpointing semantics and a T5 memory leak do not share the same underlying defect."
5279
  },
5280
  {
5281
  "left": "issue:42886",
5282
  "right": "issue:44843",
5283
+ "accept": false,
5284
+ "reason": "Tokenizer cache loading in offline mode and an unconditional model_info() call in mistral regex patching are related to offline behavior but are still different failing paths."
5285
  },
5286
  {
5287
  "left": "issue:45759",
5288
  "right": "issue:45923",
5289
  "accept": false,
5290
+ "reason": "from_config unwrapping for composite configs is separate from missing gradient-checkpointing support on a remote-code model."
5291
  },
5292
  {
5293
  "left": "issue:44514",
5294
  "right": "issue:44545",
5295
  "accept": true,
5296
+ "reason": "Same crash scenario and essentially the same title/body: apply_chat_template on batched input with padding=False."
5297
  },
5298
  {
5299
  "left": "issue:46032",
5300
  "right": "issue:46463",
5301
  "accept": false,
5302
+ "reason": "Mamba2Mixer cache correctness and T5 decoder-input validation messaging are unrelated bugs."
5303
  },
5304
  {
5305
  "left": "issue:39692",
5306
  "right": "issue:43873",
5307
  "accept": false,
5308
+ "reason": "A SigLIP2 doc/example issue is not the same as quantization offloading not working as expected."
5309
  },
5310
  {
5311
  "left": "issue:43746",
5312
  "right": "issue:43873",
5313
  "accept": false,
5314
+ "reason": "Local checkpoint loading with PEFT and quantization offloading regressions are different code paths."
5315
  },
5316
  {
5317
  "left": "issue:43994",
5318
  "right": "issue:44671",
5319
  "accept": false,
5320
+ "reason": "SigLIP2 nonsensical outputs and CamemBERT masked-LM regression are distinct model-specific problems."
5321
  },
5322
  {
5323
  "left": "issue:43421",
5324
  "right": "issue:43653",
5325
  "accept": false,
5326
+ "reason": "Updating tokenizer post-processors at runtime is different from BigBirdTokenizer special-token registration causing empty decode output."
5327
  },
5328
  {
5329
  "left": "issue:30064",
5330
  "right": "issue:36010",
5331
  "accept": false,
5332
+ "reason": "Image segmentation-map processing and a GenerationMixin import error are completely unrelated."
5333
  },
5334
  {
5335
  "left": "issue:43782",
5336
  "right": "issue:45381",
5337
  "accept": false,
5338
+ "reason": "Both are Qwen/VL-related, but one is a weight_only loading error and the other is a video position-id regression; not the same bug."
5339
+ },
5340
+ {
5341
+ "left": "issue:43421",
5342
+ "right": "issue:44186",
5343
+ "accept": false,
5344
+ "reason": "Special-token update behavior in TokenizersBackend is not the same as LayoutLMv2Tokenizer crashing on NER/batched padding."
5345
  },
5346
  {
5347
  "left": "issue:45081",
5348
  "right": "issue:45356",
5349
  "accept": false,
5350
+ "reason": "Mistral-regex AttributeError and Kimi-K2.5 codec/regression behavior are related to tokenizer patching, but they are not the same concrete failure."
5351
  },
5352
  {
5353
  "left": "issue:45356",
5354
  "right": "issue:45907",
5355
  "accept": false,
5356
+ "reason": "A tokenizer regression is not the same as list_repo_files failing to forward revision; different components and fixes."
5357
  }
5358
  ]
5359
  },
5360
  "evaluator_result": {
5361
+ "accept": false,
5362
+ "feedback": "The duplicate-pair call on issue:44514/issue:44545 is plausible, but the canonicalization rationale is not fully grounded: the packet shows issue:44545 has more discussion activity and more inbound references, so saying issue:44514 has the 'stronger discussion trail' is unsupported. A few soft-edge rejections are also a bit stronger than the packet alone can prove, especially where the titles indicate related tokenizer/offline behavior."
5363
  },
5364
  "evaluator_used": true,
5365
+ "retried": true,
5366
  "accepted_nontrivial_soft_edge": true,
5367
  "error_kind": null,
5368
  "error_message": null
 
5679
  "review_unit_count": 58,
5680
  "status": "reviewed",
5681
  "reason": null,
5682
+ "source": "llm",
5683
+ "cache_hit": false,
5684
  "model": "gpt-5.4-mini?service_tier=flex",
5685
  "cluster_id": "cluster-10000-484",
5686
  "nodes": [
 
5761
  "issue:45987|issue:46192",
5762
  "issue:44671|issue:45701"
5763
  ],
5764
+ "prepared_review_unit_hash": "ac41017ebdf616560ac7d51f780e290cb0febc2e580d93f684bd95c3a0e52841",
5765
  "component_budget": {
5766
  "node_count": 484,
5767
  "item_count": 484,
 
5783
  "aggressively_trimmed": false,
5784
  "split": true,
5785
  "analyst_result": {
5786
+ "summary": "This cluster is mostly heterogeneous regression reports. The only strong duplicate pair is the tokenizer spacing/inconsistency reports (28218 and 29159); the rest are related only at a broad subsystem level or are clearly separate bugs.",
5787
+ "confidence": 0.79,
5788
+ "canonical_issue_reason": "issue:29159 is the broader, already-closed tokenizer inconsistency report and best fits the more specific added-token spacing symptom described in issue:28218.",
5789
  "canonical_pr_reason": null,
5790
+ "best_issue_reason": "issue:29159 is the best representative because it is the broadest issue title in the only clearly duplicated area, and it can subsume the narrower extra-space-after-added-token symptom.",
5791
  "best_pr_reason": null,
5792
  "soft_edge_verdicts": [
5793
  {
5794
  "left": "issue:34634",
5795
  "right": "issue:36010",
5796
  "accept": false,
5797
+ "reason": "Different components and failure modes: BarkProcessor voice_preset behavior vs a GenerationMixin import error."
5798
  },
5799
  {
5800
  "left": "issue:43606",
5801
  "right": "issue:45072",
5802
  "accept": false,
5803
+ "reason": "Both mention inference/device behavior, but one is a CPU offload device mismatch and the other is a bfloat16 dtype mismatch in a different model path."
5804
  },
5805
  {
5806
  "left": "issue:45183",
5807
  "right": "issue:45479",
5808
  "accept": false,
5809
+ "reason": "These are unrelated validation/loss bugs: input_ids/inputs_embeds XOR validation vs single-label classification with num_labels=1."
5810
  },
5811
  {
5812
  "left": "issue:45584",
5813
  "right": "issue:45834",
5814
  "accept": false,
5815
+ "reason": "Whisper empty-transcription generation after align_special_tokens is unrelated to Kosmos2.5 long OCR index errors."
5816
  },
5817
  {
5818
  "left": "issue:43716",
5819
  "right": "issue:44008",
5820
  "accept": false,
5821
+ "reason": "Different multimodal bugs: image preprocessor dtype mismatch vs Gemma3n variable-name collision causing audio_mel_mask access failure."
5822
  },
5823
  {
5824
  "left": "issue:43950",
5825
  "right": "issue:45987",
5826
  "accept": false,
5827
+ "reason": "One is silent corruption of non-persistent buffers in from_pretrained; the other is StaticCache returning the wrong type."
5828
  },
5829
  {
5830
  "left": "issue:44479",
5831
  "right": "issue:45834",
5832
  "accept": false,
5833
+ "reason": "Different model families and symptoms: a Qwen video regression vs Kosmos2.5 OCR index error."
5834
  },
5835
  {
5836
  "left": "issue:43582",
5837
  "right": "issue:46512",
5838
  "accept": false,
5839
+ "reason": "Apple Silicon warmup TypeError is unrelated to from_config leaving config.dtype stale."
5840
  },
5841
  {
5842
  "left": "issue:43388",
5843
  "right": "issue:46032",
5844
  "accept": false,
5845
+ "reason": "Different training/utilities bugs: gather_for_metrics label dropping vs Mamba2 use_cache producing incorrect outputs."
5846
  },
5847
  {
5848
  "left": "issue:44991",
5849
  "right": "issue:46291",
5850
  "accept": false,
5851
+ "reason": "Tokenizer loading regression and pipeline import failure are separate breakages with different code paths."
5852
  },
5853
  {
5854
  "left": "issue:43577",
5855
  "right": "issue:43818",
5856
  "accept": false,
5857
+ "reason": "BLIP2 dtype persistence and Video-LLaVA temporal-attention/weight-sharing issues are unrelated."
5858
  },
5859
  {
5860
  "left": "issue:43825",
5861
  "right": "issue:43994",
5862
  "accept": false,
5863
+ "reason": "One is an error-message wording issue; the other is a model/pipeline result correctness issue."
5864
  },
5865
  {
5866
  "left": "issue:46294",
5867
  "right": "issue:46463",
5868
  "accept": false,
5869
+ "reason": "Qwen3.5 video token/feature mismatch is unrelated to T5 decoder input validation messaging."
5870
  },
5871
  {
5872
  "left": "issue:43782",
5873
  "right": "issue:44977",
5874
  "accept": false,
5875
+ "reason": "Different generation/loading problems: weight_only loading error vs flash-attention generation regression."
5876
  },
5877
  {
5878
  "left": "issue:44561",
5879
  "right": "issue:45020",
5880
  "accept": false,
5881
+ "reason": "The first is a specific missing utility breaking trust_remote_code models; the second is a broad umbrella report that could cover many regressions, so they are not the same concrete bug."
5882
  },
5883
  {
5884
  "left": "issue:43421",
5885
  "right": "issue:44811",
5886
  "accept": false,
5887
+ "reason": "Special-token post-processor updates and Whisper batch_decode skip_special_tokens handling are distinct tokenizer/processor behaviors."
5888
  },
5889
  {
5890
  "left": "issue:44805",
5891
  "right": "issue:44936",
5892
  "accept": false,
5893
+ "reason": "A mask/index shape error is not the same bug as trainer.evaluate() failing after trainer.train()."
5894
  },
5895
  {
5896
  "left": "issue:44610",
5897
  "right": "issue:45072",
5898
  "accept": false,
5899
+ "reason": "Processor image-size mismatch and bfloat16 dtype mismatches are unrelated."
5900
  },
5901
  {
5902
  "left": "issue:43720",
5903
  "right": "issue:45072",
5904
  "accept": false,
5905
+ "reason": "Packed-weight unloading during accelerate loading is unrelated to dtype mismatch inference bugs."
5906
  },
5907
  {
5908
  "left": "issue:43540",
5909
  "right": "issue:45381",
5910
  "accept": false,
5911
+ "reason": "Both involve video inputs, but they target different Qwen model families and different failure mechanisms."
5912
  },
5913
  {
5914
  "left": "issue:43825",
5915
  "right": "issue:43827",
5916
  "accept": false,
5917
+ "reason": "Both reference pipeline removal fallout, but one is an error message and the other is stale docs; not the same concrete bug."
5918
  },
5919
  {
5920
  "left": "issue:28218",
5921
  "right": "issue:29159",
5922
  "accept": true,
5923
+ "reason": "These describe the same tokenizer inconsistency: a specific extra-space symptom after added tokens versus the broader slow/fast tokenizer mismatch."
5924
  },
5925
  {
5926
  "left": "issue:43749",
5927
  "right": "issue:44291",
5928
  "accept": false,
5929
+ "reason": "Low-memory loading regressions, but one is FSDP_CPU_RAM_EFFICIENT_LOADING and the other is init_empty_weights argument handling."
5930
  },
5931
  {
5932
  "left": "issue:43645",
5933
  "right": "issue:46665",
5934
  "accept": false,
5935
+ "reason": "Jupyter custom-model initialization breakage is not the same as inconsistent from_pretrained behavior inside Python classes."
5936
  },
5937
  {
5938
  "left": "issue:43577",
5939
  "right": "issue:43582",
5940
  "accept": false,
5941
+ "reason": "BLIP2 dtype mismatch and Apple Silicon allocator warmup TypeError are unrelated."
5942
  },
5943
  {
5944
  "left": "issue:45987",
5945
  "right": "issue:46192",
5946
  "accept": false,
5947
+ "reason": "A cache length return-type bug is unrelated to return_tensors being ignored when text_kwargs is passed."
5948
  },
5949
  {
5950
  "left": "issue:44671",
5951
  "right": "issue:45701",
5952
  "accept": false,
5953
+ "reason": "CamemBERT masked-LM correctness in v5 and a broad tokenization-version change are too different to treat as the same bug."
5954
+ },
5955
+ {
5956
+ "left": "issue:43577",
5957
+ "right": "issue:43818",
5958
+ "accept": false,
5959
+ "reason": "Repeated pair: unrelated dtype-loading issue vs Video-LLaVA architecture/weight-sharing issue."
5960
+ },
5961
+ {
5962
+ "left": "issue:46294",
5963
+ "right": "issue:46463",
5964
+ "accept": false,
5965
+ "reason": "Repeated pair: Qwen video feature/token mismatch is unrelated to T5 decoder validation messaging."
5966
  }
5967
  ]
5968
  },
5969
  "evaluator_result": {
5970
  "accept": true,
5971
+ "feedback": "The summary is grounded in the packet: 28218/29159 is the only clearly supported duplicate-style pair, and the rest of the soft-edge verdicts stay conservative by distinguishing different components, model families, or failure modes. No overstatement is apparent."
5972
  },
5973
  "evaluator_used": true,
5974
  "retried": false,
 
8105
  "review_unit_count": 58,
8106
  "status": "reviewed",
8107
  "reason": null,
8108
+ "source": "cache",
8109
+ "cache_hit": true,
8110
  "model": "gpt-5.4-mini?service_tier=flex",
8111
  "cluster_id": "cluster-10000-484",
8112
  "nodes": [
 
11115
  "review_unit_count": 58,
11116
  "status": "reviewed",
11117
  "reason": null,
11118
+ "source": "cache",
11119
+ "cache_hit": true,
11120
  "model": "gpt-5.4-mini?service_tier=flex",
11121
  "cluster_id": "cluster-10000-484",
11122
  "nodes": [
 
11705
  "review_unit_count": 58,
11706
  "status": "reviewed",
11707
  "reason": null,
11708
+ "source": "cache",
11709
+ "cache_hit": true,
11710
  "model": "gpt-5.4-mini?service_tier=flex",
11711
  "cluster_id": "cluster-10000-484",
11712
  "nodes": [
 
12320
  "review_unit_count": 58,
12321
  "status": "reviewed",
12322
  "reason": null,
12323
+ "source": "cache",
12324
+ "cache_hit": true,
12325
  "model": "gpt-5.4-mini?service_tier=flex",
12326
  "cluster_id": "cluster-10000-484",
12327
  "nodes": [
 
12625
  "review_unit_count": 58,
12626
  "status": "reviewed",
12627
  "reason": null,
12628
+ "source": "cache",
12629
+ "cache_hit": true,
12630
  "model": "gpt-5.4-mini?service_tier=flex",
12631
  "cluster_id": "cluster-10000-484",
12632
  "nodes": [
 
13242
  "review_unit_count": 58,
13243
  "status": "reviewed",
13244
  "reason": null,
13245
+ "source": "cache",
13246
+ "cache_hit": true,
13247
  "model": "gpt-5.4-mini?service_tier=flex",
13248
  "cluster_id": "cluster-10000-484",
13249
  "nodes": [
 
15076
  "review_unit_count": 58,
15077
  "status": "reviewed",
15078
  "reason": null,
15079
+ "source": "cache",
15080
+ "cache_hit": true,
15081
  "model": "gpt-5.4-mini?service_tier=flex",
15082
  "cluster_id": "cluster-10000-484",
15083
  "nodes": [
 
16039
  "review_unit_count": 58,
16040
  "status": "reviewed",
16041
  "reason": null,
16042
+ "source": "cache",
16043
+ "cache_hit": true,
16044
  "model": "gpt-5.4-mini?service_tier=flex",
16045
  "cluster_id": "cluster-10000-484",
16046
  "nodes": [
 
20330
  "review_unit_count": 1,
20331
  "status": "reviewed",
20332
  "reason": null,
20333
+ "source": "cache",
20334
+ "cache_hit": true,
20335
  "model": "gpt-5.4-mini?service_tier=flex",
20336
  "cluster_id": "cluster-40546-2",
20337
  "nodes": [
 
20393
  "review_unit_count": 1,
20394
  "status": "reviewed",
20395
  "reason": null,
20396
+ "source": "cache",
20397
+ "cache_hit": true,
20398
  "model": "gpt-5.4-mini?service_tier=flex",
20399
  "cluster_id": "cluster-40755-2",
20400
  "nodes": [
 
20519
  "review_unit_count": 1,
20520
  "status": "reviewed",
20521
  "reason": null,
20522
+ "source": "cache",
20523
+ "cache_hit": true,
20524
  "model": "gpt-5.4-mini?service_tier=flex",
20525
  "cluster_id": "cluster-41855-14",
20526
  "nodes": [
 
20723
  "review_unit_count": 1,
20724
  "status": "reviewed",
20725
  "reason": null,
20726
+ "source": "cache",
20727
+ "cache_hit": true,
20728
  "model": "gpt-5.4-mini?service_tier=flex",
20729
  "cluster_id": "cluster-42975-2",
20730
  "nodes": [
 
20786
  "review_unit_count": 1,
20787
  "status": "reviewed",
20788
  "reason": null,
20789
+ "source": "cache",
20790
+ "cache_hit": true,
20791
  "model": "gpt-5.4-mini?service_tier=flex",
20792
  "cluster_id": "cluster-42978-7",
20793
  "nodes": [
 
20931
  "review_unit_count": 1,
20932
  "status": "reviewed",
20933
  "reason": null,
20934
+ "source": "cache",
20935
+ "cache_hit": true,
20936
  "model": "gpt-5.4-mini?service_tier=flex",
20937
  "cluster_id": "cluster-43291-8",
20938
  "nodes": [
 
21091
  "review_unit_count": 1,
21092
  "status": "reviewed",
21093
  "reason": null,
21094
+ "source": "cache",
21095
+ "cache_hit": true,
21096
  "model": "gpt-5.4-mini?service_tier=flex",
21097
  "cluster_id": "cluster-43488-3",
21098
  "nodes": [
 
21169
  "review_unit_count": 1,
21170
  "status": "reviewed",
21171
  "reason": null,
21172
+ "source": "cache",
21173
+ "cache_hit": true,
21174
  "model": "gpt-5.4-mini?service_tier=flex",
21175
  "cluster_id": "cluster-43612-2",
21176
  "nodes": [
 
21232
  "review_unit_count": 1,
21233
  "status": "reviewed",
21234
  "reason": null,
21235
+ "source": "cache",
21236
+ "cache_hit": true,
21237
  "model": "gpt-5.4-mini?service_tier=flex",
21238
  "cluster_id": "cluster-43651-2",
21239
  "nodes": [
 
21295
  "review_unit_count": 1,
21296
  "status": "reviewed",
21297
  "reason": null,
21298
+ "source": "cache",
21299
+ "cache_hit": true,
21300
  "model": "gpt-5.4-mini?service_tier=flex",
21301
  "cluster_id": "cluster-43656-2",
21302
  "nodes": [
 
21424
  "review_unit_count": 2,
21425
  "status": "reviewed",
21426
  "reason": null,
21427
+ "source": "cache",
21428
+ "cache_hit": true,
21429
  "model": "gpt-5.4-mini?service_tier=flex",
21430
  "cluster_id": "cluster-43996-28",
21431
  "nodes": [
 
21910
  "review_unit_count": 2,
21911
  "status": "reviewed",
21912
  "reason": null,
21913
+ "source": "cache",
21914
+ "cache_hit": true,
21915
  "model": "gpt-5.4-mini?service_tier=flex",
21916
  "cluster_id": "cluster-43996-28",
21917
  "nodes": [
 
22014
  "review_unit_count": 1,
22015
  "status": "reviewed",
22016
  "reason": null,
22017
+ "source": "cache",
22018
+ "cache_hit": true,
22019
  "model": "gpt-5.4-mini?service_tier=flex",
22020
  "cluster_id": "cluster-43998-13",
22021
  "nodes": [
 
22355
  "review_unit_count": 1,
22356
  "status": "reviewed",
22357
  "reason": null,
22358
+ "source": "cache",
22359
+ "cache_hit": true,
22360
  "model": "gpt-5.4-mini?service_tier=flex",
22361
  "cluster_id": "cluster-44662-2",
22362
  "nodes": [
 
22418
  "review_unit_count": 1,
22419
  "status": "reviewed",
22420
  "reason": null,
22421
+ "source": "cache",
22422
+ "cache_hit": true,
22423
  "model": "gpt-5.4-mini?service_tier=flex",
22424
  "cluster_id": "cluster-44979-2",
22425
  "nodes": [
 
22481
  "review_unit_count": 1,
22482
  "status": "reviewed",
22483
  "reason": null,
22484
+ "source": "cache",
22485
+ "cache_hit": true,
22486
  "model": "gpt-5.4-mini?service_tier=flex",
22487
  "cluster_id": "cluster-45086-2",
22488
  "nodes": [
 
22544
  "review_unit_count": 1,
22545
  "status": "reviewed",
22546
  "reason": null,
22547
+ "source": "cache",
22548
+ "cache_hit": true,
22549
  "model": "gpt-5.4-mini?service_tier=flex",
22550
  "cluster_id": "cluster-45115-3",
22551
  "nodes": [
 
22615
  "review_unit_count": 1,
22616
  "status": "reviewed",
22617
  "reason": null,
22618
+ "source": "cache",
22619
+ "cache_hit": true,
22620
  "model": "gpt-5.4-mini?service_tier=flex",
22621
  "cluster_id": "cluster-45332-2",
22622
  "nodes": [
 
22678
  "review_unit_count": 1,
22679
  "status": "reviewed",
22680
  "reason": null,
22681
+ "source": "cache",
22682
+ "cache_hit": true,
22683
  "model": "gpt-5.4-mini?service_tier=flex",
22684
  "cluster_id": "cluster-45415-2",
22685
  "nodes": [
 
22752
  "soft_pairs": [
22753
  "pull_request:45513|pull_request:46084"
22754
  ],
22755
+ "prepared_review_unit_hash": "30545ea611f2077bd17416dfd71c1d6d41679b4c176efa83feec8a9aa2c17308",
22756
  "component_budget": {
22757
  "node_count": 2,
22758
  "item_count": 2,
 
22774
  "aggressively_trimmed": false,
22775
  "split": false,
22776
  "analyst_result": {
22777
+ "summary": "These are unrelated pull requests: one fixes multi-token cached forward for Qwen3.5/related linear-attention models, the other fixes use_cache behavior for Mamba2 when seq_len > 1. They touch different code paths and model families, so they should not be clustered as duplicates.",
22778
  "confidence": 0.98,
22779
  "canonical_issue_reason": null,
22780
+ "canonical_pr_reason": "PR #46084 is the more canonical representative because it targets a single concrete bug with an explicit issue link and a focused seq_len > 1 cache-path fix in Mamba2.",
22781
  "best_issue_reason": null,
22782
+ "best_pr_reason": "PR #46084 is the best representative for the cluster due to its explicit issue target and narrowly scoped fix; PR #45513 is broader and covers multiple model files, but a different bug.",
22783
  "soft_edge_verdicts": [
22784
  {
22785
  "left": "pull_request:45513",
22786
  "right": "pull_request:46084",
22787
  "accept": false,
22788
+ "reason": "Different model families and different failure modes: Qwen3.5/OLMo/Qwen3-next multi-token cached forward vs Mamba2 use_cache with seq_len > 1. They are not the same underlying bug and would not plausibly merge into one PR."
22789
  }
22790
  ]
22791
  },
 
22804
  "review_unit_count": 1,
22805
  "status": "reviewed",
22806
  "reason": null,
22807
+ "source": "cache",
22808
+ "cache_hit": true,
22809
  "model": "gpt-5.4-mini?service_tier=flex",
22810
  "cluster_id": "cluster-45548-2",
22811
  "nodes": [
 
22867
  "review_unit_count": 1,
22868
  "status": "reviewed",
22869
  "reason": null,
22870
+ "source": "cache",
22871
+ "cache_hit": true,
22872
  "model": "gpt-5.4-mini?service_tier=flex",
22873
  "cluster_id": "cluster-45639-2",
22874
  "nodes": [
 
22930
  "review_unit_count": 1,
22931
  "status": "reviewed",
22932
  "reason": null,
22933
+ "source": "cache",
22934
+ "cache_hit": true,
22935
  "model": "gpt-5.4-mini?service_tier=flex",
22936
  "cluster_id": "cluster-45720-2",
22937
  "nodes": [
 
23059
  "review_unit_count": 1,
23060
  "status": "reviewed",
23061
  "reason": null,
23062
+ "source": "cache",
23063
+ "cache_hit": true,
23064
  "model": "gpt-5.4-mini?service_tier=flex",
23065
  "cluster_id": "cluster-45932-2",
23066
  "nodes": [
 
23122
  "review_unit_count": 1,
23123
  "status": "reviewed",
23124
  "reason": null,
23125
+ "source": "cache",
23126
+ "cache_hit": true,
23127
  "model": "gpt-5.4-mini?service_tier=flex",
23128
  "cluster_id": "cluster-46126-2",
23129
  "nodes": [
 
23185
  "review_unit_count": 1,
23186
  "status": "reviewed",
23187
  "reason": null,
23188
+ "source": "cache",
23189
+ "cache_hit": true,
23190
  "model": "gpt-5.4-mini?service_tier=flex",
23191
  "cluster_id": "cluster-46138-10",
23192
  "nodes": [
 
23452
  "review_unit_count": 1,
23453
  "status": "reviewed",
23454
  "reason": null,
23455
+ "source": "cache",
23456
+ "cache_hit": true,
23457
  "model": "gpt-5.4-mini?service_tier=flex",
23458
  "cluster_id": "cluster-46152-2",
23459
  "nodes": [
 
23515
  "review_unit_count": 1,
23516
  "status": "reviewed",
23517
  "reason": null,
23518
+ "source": "cache",
23519
+ "cache_hit": true,
23520
  "model": "gpt-5.4-mini?service_tier=flex",
23521
  "cluster_id": "cluster-46154-2",
23522
  "nodes": [
 
23707
  "review_unit_count": 1,
23708
  "status": "reviewed",
23709
  "reason": null,
23710
+ "source": "cache",
23711
+ "cache_hit": true,
23712
  "model": "gpt-5.4-mini?service_tier=flex",
23713
  "cluster_id": "cluster-46310-2",
23714
  "nodes": [
 
23770
  "review_unit_count": 1,
23771
  "status": "reviewed",
23772
  "reason": null,
23773
+ "source": "cache",
23774
+ "cache_hit": true,
23775
  "model": "gpt-5.4-mini?service_tier=flex",
23776
  "cluster_id": "cluster-46376-2",
23777
  "nodes": [
 
23833
  "review_unit_count": 1,
23834
  "status": "reviewed",
23835
  "reason": null,
23836
+ "source": "cache",
23837
+ "cache_hit": true,
23838
  "model": "gpt-5.4-mini?service_tier=flex",
23839
  "cluster_id": "cluster-46574-2",
23840
  "nodes": [
 
23896
  "review_unit_count": 1,
23897
  "status": "reviewed",
23898
  "reason": null,
23899
+ "source": "cache",
23900
+ "cache_hit": true,
23901
  "model": "gpt-5.4-mini?service_tier=flex",
23902
  "cluster_id": "cluster-46603-2",
23903
  "nodes": [
 
23976
  "pull_request:46748|pull_request:46781",
23977
  "pull_request:46748|pull_request:46773"
23978
  ],
23979
+ "prepared_review_unit_hash": "dadee70a398944009af4319f4303489824a7d2b7f01ebf9280da9b4f04f9c4ee",
23980
  "component_budget": {
23981
  "node_count": 4,
23982
  "item_count": 4,
 
23998
  "aggressively_trimmed": false,
23999
  "split": false,
24000
  "analyst_result": {
24001
+ "summary": "These PRs are related by hybrid/linear-attention model infrastructure, but they fix distinct bugs in different code paths. The closest thematic overlap is between the two cache_utils PRs, yet they are not duplicates.",
24002
+ "confidence": 0.88,
24003
  "canonical_issue_reason": null,
24004
+ "canonical_pr_reason": "PR 46748 is the closest umbrella representative: it is a concrete cache_utils bugfix with tests and sits at the shared hybrid-model cache layer, though it does not subsume the other PRs.",
24005
  "best_issue_reason": null,
24006
+ "best_pr_reason": "PR 46748 is the best standalone candidate: the bug is specific, reproducible, and centrally located in shared infrastructure, with a clear test surface and narrow scope.",
24007
  "soft_edge_verdicts": [
24008
  {
24009
  "left": "pull_request:46741",
24010
  "right": "pull_request:46781",
24011
  "accept": false,
24012
+ "reason": "Different fixes and code paths: Nemotron/Zamba2 chunked-prefill speculative decoding vs QuantizedCache construction on hybrid models."
24013
  },
24014
  {
24015
  "left": "pull_request:46741",
24016
  "right": "pull_request:46773",
24017
  "accept": false,
24018
+ "reason": "Both touch forward logic, but one fixes chunked-prefill/speculative verify while the other fixes padding-mask application for batch size 1; not the same bug."
24019
  },
24020
  {
24021
  "left": "pull_request:46741",
24022
  "right": "pull_request:46748",
24023
  "accept": false,
24024
+ "reason": "Model-specific mixer logic bug vs cache offload prefetch device mismatch; unrelated failure modes."
24025
  },
24026
  {
24027
  "left": "pull_request:46748",
24028
+ "right": "pull_request:46781",
24029
  "accept": false,
24030
+ "reason": "Both are cache_utils/hybrid-model fixes, but one addresses offloaded cache prefetching and the other QuantizedCache layer-type selection; different concrete code paths."
24031
  },
24032
  {
24033
  "left": "pull_request:46748",
24034
+ "right": "pull_request:46773",
24035
  "accept": false,
24036
+ "reason": "Cache offload bug vs padding-mask bug in linear-attention models; no shared underlying defect."
24037
  }
24038
  ]
24039
  },
 
24052
  "review_unit_count": 1,
24053
  "status": "reviewed",
24054
  "reason": null,
24055
+ "source": "cache",
24056
+ "cache_hit": true,
24057
  "model": "gpt-5.4-mini?service_tier=flex",
24058
  "cluster_id": "cluster-46753-6",
24059
  "nodes": [
 
24192
  "review_unit_count": 1,
24193
  "status": "reviewed",
24194
  "reason": null,
24195
+ "source": "cache",
24196
+ "cache_hit": true,
24197
  "model": "gpt-5.4-mini?service_tier=flex",
24198
  "cluster_id": "cluster-46757-2",
24199
  "nodes": [
analysis/current/manifest.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "analysis_id": "hybrid-model-20260621t180014z",
3
  "archived_artifacts": {
4
- "hybrid": "snapshots/20260621T180014Z/analysis-runs/hybrid-model-20260621t180014z/analysis-report-hybrid.json",
5
- "hybrid_reviews": "snapshots/20260621T180014Z/analysis-runs/hybrid-model-20260621t180014z/analysis-report-hybrid.llm-reviews.json"
6
  },
7
  "artifacts": {
8
  "hybrid": "analysis/current/analysis-report-hybrid.json",
@@ -10,9 +10,9 @@
10
  },
11
  "channel": "canonical",
12
  "model": null,
13
- "published_at": "2026-06-21T18:06:02Z",
14
  "repo": "huggingface/transformers",
15
  "schema_version": 1,
16
- "snapshot_id": "20260621T180014Z",
17
  "variant": "hybrid"
18
  }
 
1
  {
2
+ "analysis_id": "hybrid-model-20260622t000017z",
3
  "archived_artifacts": {
4
+ "hybrid": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.json",
5
+ "hybrid_reviews": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.llm-reviews.json"
6
  },
7
  "artifacts": {
8
  "hybrid": "analysis/current/analysis-report-hybrid.json",
 
10
  },
11
  "channel": "canonical",
12
  "model": null,
13
+ "published_at": "2026-06-22T00:04:30Z",
14
  "repo": "huggingface/transformers",
15
  "schema_version": 1,
16
+ "snapshot_id": "20260622T000017Z",
17
  "variant": "hybrid"
18
  }
snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.json ADDED
@@ -0,0 +1,3039 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "1.0",
3
+ "repo": "huggingface/transformers",
4
+ "snapshot_id": "20260622T000017Z",
5
+ "generated_at": "2026-06-22T00:04:29Z",
6
+ "evidence_quality": "full",
7
+ "llm_enrichment": true,
8
+ "meta_bugs": [
9
+ {
10
+ "cluster_id": "cluster-43979-11",
11
+ "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.",
12
+ "status": "open",
13
+ "confidence": 0.8,
14
+ "canonical_issue_number": 43979,
15
+ "canonical_pr_number": 44007,
16
+ "issue_numbers": [
17
+ 43979
18
+ ],
19
+ "pr_numbers": [
20
+ 43996,
21
+ 44007,
22
+ 44013,
23
+ 44044,
24
+ 44066,
25
+ 44072,
26
+ 44085,
27
+ 44129,
28
+ 44154,
29
+ 44722
30
+ ],
31
+ "evidence_types": [
32
+ "closing_reference",
33
+ "shared_issue_target",
34
+ "soft_similarity"
35
+ ],
36
+ "pr_comparisons": [
37
+ {
38
+ "left_pr_number": 43996,
39
+ "right_pr_number": 44007,
40
+ "code_similarity": 0.179,
41
+ "size_similarity": 0.576,
42
+ "file_overlap": 0.0,
43
+ "area_overlap": 0.0,
44
+ "patch_similarity": 0.429,
45
+ "shared_filenames": [],
46
+ "shared_file_areas": []
47
+ },
48
+ {
49
+ "left_pr_number": 43996,
50
+ "right_pr_number": 44013,
51
+ "code_similarity": 0.122,
52
+ "size_similarity": 0.318,
53
+ "file_overlap": 0.0,
54
+ "area_overlap": 0.0,
55
+ "patch_similarity": 0.392,
56
+ "shared_filenames": [],
57
+ "shared_file_areas": []
58
+ },
59
+ {
60
+ "left_pr_number": 43996,
61
+ "right_pr_number": 44044,
62
+ "code_similarity": 0.245,
63
+ "size_similarity": 0.864,
64
+ "file_overlap": 0.0,
65
+ "area_overlap": 0.0,
66
+ "patch_similarity": 0.479,
67
+ "shared_filenames": [],
68
+ "shared_file_areas": []
69
+ },
70
+ {
71
+ "left_pr_number": 43996,
72
+ "right_pr_number": 44066,
73
+ "code_similarity": 0.225,
74
+ "size_similarity": 0.818,
75
+ "file_overlap": 0.0,
76
+ "area_overlap": 0.0,
77
+ "patch_similarity": 0.408,
78
+ "shared_filenames": [],
79
+ "shared_file_areas": []
80
+ },
81
+ {
82
+ "left_pr_number": 43996,
83
+ "right_pr_number": 44072,
84
+ "code_similarity": 0.14,
85
+ "size_similarity": 0.303,
86
+ "file_overlap": 0.0,
87
+ "area_overlap": 0.0,
88
+ "patch_similarity": 0.528,
89
+ "shared_filenames": [],
90
+ "shared_file_areas": []
91
+ },
92
+ {
93
+ "left_pr_number": 43996,
94
+ "right_pr_number": 44085,
95
+ "code_similarity": 0.216,
96
+ "size_similarity": 0.783,
97
+ "file_overlap": 0.0,
98
+ "area_overlap": 0.0,
99
+ "patch_similarity": 0.398,
100
+ "shared_filenames": [],
101
+ "shared_file_areas": []
102
+ },
103
+ {
104
+ "left_pr_number": 43996,
105
+ "right_pr_number": 44129,
106
+ "code_similarity": 0.163,
107
+ "size_similarity": 0.643,
108
+ "file_overlap": 0.0,
109
+ "area_overlap": 0.0,
110
+ "patch_similarity": 0.229,
111
+ "shared_filenames": [],
112
+ "shared_file_areas": []
113
+ },
114
+ {
115
+ "left_pr_number": 43996,
116
+ "right_pr_number": 44154,
117
+ "code_similarity": 0.153,
118
+ "size_similarity": 0.535,
119
+ "file_overlap": 0.0,
120
+ "area_overlap": 0.0,
121
+ "patch_similarity": 0.31,
122
+ "shared_filenames": [],
123
+ "shared_file_areas": []
124
+ },
125
+ {
126
+ "left_pr_number": 43996,
127
+ "right_pr_number": 44722,
128
+ "code_similarity": 0.225,
129
+ "size_similarity": 0.848,
130
+ "file_overlap": 0.0,
131
+ "area_overlap": 0.0,
132
+ "patch_similarity": 0.368,
133
+ "shared_filenames": [],
134
+ "shared_file_areas": []
135
+ },
136
+ {
137
+ "left_pr_number": 44007,
138
+ "right_pr_number": 44013,
139
+ "code_similarity": 0.19,
140
+ "size_similarity": 0.553,
141
+ "file_overlap": 0.0,
142
+ "area_overlap": 0.0,
143
+ "patch_similarity": 0.531,
144
+ "shared_filenames": [],
145
+ "shared_file_areas": []
146
+ },
147
+ {
148
+ "left_pr_number": 44007,
149
+ "right_pr_number": 44044,
150
+ "code_similarity": 0.186,
151
+ "size_similarity": 0.667,
152
+ "file_overlap": 0.0,
153
+ "area_overlap": 0.0,
154
+ "patch_similarity": 0.354,
155
+ "shared_filenames": [],
156
+ "shared_file_areas": []
157
+ },
158
+ {
159
+ "left_pr_number": 44007,
160
+ "right_pr_number": 44066,
161
+ "code_similarity": 0.188,
162
+ "size_similarity": 0.704,
163
+ "file_overlap": 0.0,
164
+ "area_overlap": 0.0,
165
+ "patch_similarity": 0.315,
166
+ "shared_filenames": [],
167
+ "shared_file_areas": []
168
+ },
169
+ {
170
+ "left_pr_number": 44007,
171
+ "right_pr_number": 44072,
172
+ "code_similarity": 0.212,
173
+ "size_similarity": 0.526,
174
+ "file_overlap": 0.0,
175
+ "area_overlap": 0.0,
176
+ "patch_similarity": 0.708,
177
+ "shared_filenames": [],
178
+ "shared_file_areas": []
179
+ },
180
+ {
181
+ "left_pr_number": 44007,
182
+ "right_pr_number": 44085,
183
+ "code_similarity": 0.195,
184
+ "size_similarity": 0.735,
185
+ "file_overlap": 0.0,
186
+ "area_overlap": 0.0,
187
+ "patch_similarity": 0.318,
188
+ "shared_filenames": [],
189
+ "shared_file_areas": []
190
+ },
191
+ {
192
+ "left_pr_number": 44007,
193
+ "right_pr_number": 44129,
194
+ "code_similarity": 0.103,
195
+ "size_similarity": 0.37,
196
+ "file_overlap": 0.0,
197
+ "area_overlap": 0.0,
198
+ "patch_similarity": 0.191,
199
+ "shared_filenames": [],
200
+ "shared_file_areas": []
201
+ },
202
+ {
203
+ "left_pr_number": 44007,
204
+ "right_pr_number": 44154,
205
+ "code_similarity": 0.238,
206
+ "size_similarity": 0.93,
207
+ "file_overlap": 0.0,
208
+ "area_overlap": 0.0,
209
+ "patch_similarity": 0.344,
210
+ "shared_filenames": [],
211
+ "shared_file_areas": []
212
+ },
213
+ {
214
+ "left_pr_number": 44007,
215
+ "right_pr_number": 44722,
216
+ "code_similarity": 0.178,
217
+ "size_similarity": 0.679,
218
+ "file_overlap": 0.0,
219
+ "area_overlap": 0.0,
220
+ "patch_similarity": 0.28,
221
+ "shared_filenames": [],
222
+ "shared_file_areas": []
223
+ },
224
+ {
225
+ "left_pr_number": 44013,
226
+ "right_pr_number": 44044,
227
+ "code_similarity": 0.126,
228
+ "size_similarity": 0.368,
229
+ "file_overlap": 0.0,
230
+ "area_overlap": 0.0,
231
+ "patch_similarity": 0.351,
232
+ "shared_filenames": [],
233
+ "shared_file_areas": []
234
+ },
235
+ {
236
+ "left_pr_number": 44013,
237
+ "right_pr_number": 44066,
238
+ "code_similarity": 0.127,
239
+ "size_similarity": 0.389,
240
+ "file_overlap": 0.0,
241
+ "area_overlap": 0.0,
242
+ "patch_similarity": 0.325,
243
+ "shared_filenames": [],
244
+ "shared_file_areas": []
245
+ },
246
+ {
247
+ "left_pr_number": 44013,
248
+ "right_pr_number": 44072,
249
+ "code_similarity": 0.29,
250
+ "size_similarity": 0.952,
251
+ "file_overlap": 0.0,
252
+ "area_overlap": 0.0,
253
+ "patch_similarity": 0.667,
254
+ "shared_filenames": [],
255
+ "shared_file_areas": []
256
+ },
257
+ {
258
+ "left_pr_number": 44013,
259
+ "right_pr_number": 44085,
260
+ "code_similarity": 0.131,
261
+ "size_similarity": 0.406,
262
+ "file_overlap": 0.0,
263
+ "area_overlap": 0.0,
264
+ "patch_similarity": 0.329,
265
+ "shared_filenames": [],
266
+ "shared_file_areas": []
267
+ },
268
+ {
269
+ "left_pr_number": 44013,
270
+ "right_pr_number": 44129,
271
+ "code_similarity": 0.07,
272
+ "size_similarity": 0.205,
273
+ "file_overlap": 0.0,
274
+ "area_overlap": 0.0,
275
+ "patch_similarity": 0.192,
276
+ "shared_filenames": [],
277
+ "shared_file_areas": []
278
+ },
279
+ {
280
+ "left_pr_number": 44013,
281
+ "right_pr_number": 44154,
282
+ "code_similarity": 0.177,
283
+ "size_similarity": 0.594,
284
+ "file_overlap": 0.0,
285
+ "area_overlap": 0.0,
286
+ "patch_similarity": 0.389,
287
+ "shared_filenames": [],
288
+ "shared_file_areas": []
289
+ },
290
+ {
291
+ "left_pr_number": 44013,
292
+ "right_pr_number": 44722,
293
+ "code_similarity": 0.118,
294
+ "size_similarity": 0.375,
295
+ "file_overlap": 0.0,
296
+ "area_overlap": 0.0,
297
+ "patch_similarity": 0.287,
298
+ "shared_filenames": [],
299
+ "shared_file_areas": []
300
+ },
301
+ {
302
+ "left_pr_number": 44044,
303
+ "right_pr_number": 44066,
304
+ "code_similarity": 0.25,
305
+ "size_similarity": 0.947,
306
+ "file_overlap": 0.0,
307
+ "area_overlap": 0.0,
308
+ "patch_similarity": 0.404,
309
+ "shared_filenames": [],
310
+ "shared_file_areas": []
311
+ },
312
+ {
313
+ "left_pr_number": 44044,
314
+ "right_pr_number": 44072,
315
+ "code_similarity": 0.136,
316
+ "size_similarity": 0.351,
317
+ "file_overlap": 0.0,
318
+ "area_overlap": 0.0,
319
+ "patch_similarity": 0.442,
320
+ "shared_filenames": [],
321
+ "shared_file_areas": []
322
+ },
323
+ {
324
+ "left_pr_number": 44044,
325
+ "right_pr_number": 44085,
326
+ "code_similarity": 0.24,
327
+ "size_similarity": 0.906,
328
+ "file_overlap": 0.0,
329
+ "area_overlap": 0.0,
330
+ "patch_similarity": 0.394,
331
+ "shared_filenames": [],
332
+ "shared_file_areas": []
333
+ },
334
+ {
335
+ "left_pr_number": 44044,
336
+ "right_pr_number": 44129,
337
+ "code_similarity": 0.147,
338
+ "size_similarity": 0.555,
339
+ "file_overlap": 0.0,
340
+ "area_overlap": 0.0,
341
+ "patch_similarity": 0.243,
342
+ "shared_filenames": [],
343
+ "shared_file_areas": []
344
+ },
345
+ {
346
+ "left_pr_number": 44044,
347
+ "right_pr_number": 44154,
348
+ "code_similarity": 0.17,
349
+ "size_similarity": 0.62,
350
+ "file_overlap": 0.0,
351
+ "area_overlap": 0.0,
352
+ "patch_similarity": 0.306,
353
+ "shared_filenames": [],
354
+ "shared_file_areas": []
355
+ },
356
+ {
357
+ "left_pr_number": 44044,
358
+ "right_pr_number": 44722,
359
+ "code_similarity": 0.257,
360
+ "size_similarity": 0.982,
361
+ "file_overlap": 0.0,
362
+ "area_overlap": 0.0,
363
+ "patch_similarity": 0.402,
364
+ "shared_filenames": [],
365
+ "shared_file_areas": []
366
+ },
367
+ {
368
+ "left_pr_number": 44066,
369
+ "right_pr_number": 44072,
370
+ "code_similarity": 0.133,
371
+ "size_similarity": 0.37,
372
+ "file_overlap": 0.0,
373
+ "area_overlap": 0.0,
374
+ "patch_similarity": 0.393,
375
+ "shared_filenames": [],
376
+ "shared_file_areas": []
377
+ },
378
+ {
379
+ "left_pr_number": 44066,
380
+ "right_pr_number": 44085,
381
+ "code_similarity": 0.763,
382
+ "size_similarity": 0.957,
383
+ "file_overlap": 0.5,
384
+ "area_overlap": 0.825,
385
+ "patch_similarity": 0.887,
386
+ "shared_filenames": [
387
+ "src/transformers/models/gptj/modeling_gptj.py"
388
+ ],
389
+ "shared_file_areas": [
390
+ {
391
+ "filename": "src/transformers/models/gptj/modeling_gptj.py",
392
+ "left_ranges": [
393
+ [
394
+ 33,
395
+ 41
396
+ ],
397
+ [
398
+ 174,
399
+ 181
400
+ ],
401
+ [
402
+ 250,
403
+ 257
404
+ ],
405
+ [
406
+ 398,
407
+ 419
408
+ ],
409
+ [
410
+ 425,
411
+ 434
412
+ ],
413
+ [
414
+ 459,
415
+ 466
416
+ ],
417
+ [
418
+ 471,
419
+ 488
420
+ ],
421
+ [
422
+ 518,
423
+ 540
424
+ ],
425
+ [
426
+ 554,
427
+ 560
428
+ ],
429
+ [
430
+ 566,
431
+ 575
432
+ ],
433
+ [
434
+ 580,
435
+ 598
436
+ ],
437
+ [
438
+ 601,
439
+ 606
440
+ ],
441
+ [
442
+ 634,
443
+ 640
444
+ ],
445
+ [
446
+ 646,
447
+ 653
448
+ ],
449
+ [
450
+ 658,
451
+ 674
452
+ ],
453
+ [
454
+ 717,
455
+ 722
456
+ ],
457
+ [
458
+ 738,
459
+ 744
460
+ ],
461
+ [
462
+ 749,
463
+ 772
464
+ ],
465
+ [
466
+ 790,
467
+ 795
468
+ ]
469
+ ],
470
+ "right_ranges": [
471
+ [
472
+ 33,
473
+ 40
474
+ ],
475
+ [
476
+ 173,
477
+ 180
478
+ ],
479
+ [
480
+ 249,
481
+ 256
482
+ ],
483
+ [
484
+ 397,
485
+ 405
486
+ ],
487
+ [
488
+ 408,
489
+ 419
490
+ ],
491
+ [
492
+ 425,
493
+ 434
494
+ ],
495
+ [
496
+ 459,
497
+ 465
498
+ ],
499
+ [
500
+ 470,
501
+ 489
502
+ ],
503
+ [
504
+ 517,
505
+ 539
506
+ ],
507
+ [
508
+ 553,
509
+ 559
510
+ ],
511
+ [
512
+ 565,
513
+ 574
514
+ ],
515
+ [
516
+ 579,
517
+ 597
518
+ ],
519
+ [
520
+ 600,
521
+ 611
522
+ ],
523
+ [
524
+ 633,
525
+ 639
526
+ ],
527
+ [
528
+ 645,
529
+ 652
530
+ ],
531
+ [
532
+ 657,
533
+ 673
534
+ ],
535
+ [
536
+ 716,
537
+ 728
538
+ ],
539
+ [
540
+ 737,
541
+ 743
542
+ ],
543
+ [
544
+ 748,
545
+ 771
546
+ ],
547
+ [
548
+ 789,
549
+ 794
550
+ ]
551
+ ]
552
+ }
553
+ ]
554
+ },
555
+ {
556
+ "left_pr_number": 44066,
557
+ "right_pr_number": 44129,
558
+ "code_similarity": 0.145,
559
+ "size_similarity": 0.526,
560
+ "file_overlap": 0.0,
561
+ "area_overlap": 0.0,
562
+ "patch_similarity": 0.263,
563
+ "shared_filenames": [],
564
+ "shared_file_areas": []
565
+ },
566
+ {
567
+ "left_pr_number": 44066,
568
+ "right_pr_number": 44154,
569
+ "code_similarity": 0.174,
570
+ "size_similarity": 0.654,
571
+ "file_overlap": 0.0,
572
+ "area_overlap": 0.0,
573
+ "patch_similarity": 0.286,
574
+ "shared_filenames": [],
575
+ "shared_file_areas": []
576
+ },
577
+ {
578
+ "left_pr_number": 44066,
579
+ "right_pr_number": 44722,
580
+ "code_similarity": 0.907,
581
+ "size_similarity": 0.964,
582
+ "file_overlap": 1.0,
583
+ "area_overlap": 0.808,
584
+ "patch_similarity": 0.874,
585
+ "shared_filenames": [
586
+ "src/transformers/models/codegen/modeling_codegen.py",
587
+ "src/transformers/models/gptj/modeling_gptj.py"
588
+ ],
589
+ "shared_file_areas": [
590
+ {
591
+ "filename": "src/transformers/models/codegen/modeling_codegen.py",
592
+ "left_ranges": [
593
+ [
594
+ 245,
595
+ 266
596
+ ]
597
+ ],
598
+ "right_ranges": [
599
+ [
600
+ 228,
601
+ 234
602
+ ]
603
+ ]
604
+ },
605
+ {
606
+ "filename": "src/transformers/models/gptj/modeling_gptj.py",
607
+ "left_ranges": [
608
+ [
609
+ 33,
610
+ 41
611
+ ],
612
+ [
613
+ 174,
614
+ 181
615
+ ],
616
+ [
617
+ 250,
618
+ 257
619
+ ],
620
+ [
621
+ 398,
622
+ 419
623
+ ],
624
+ [
625
+ 425,
626
+ 434
627
+ ],
628
+ [
629
+ 459,
630
+ 466
631
+ ],
632
+ [
633
+ 471,
634
+ 488
635
+ ],
636
+ [
637
+ 518,
638
+ 540
639
+ ],
640
+ [
641
+ 554,
642
+ 560
643
+ ],
644
+ [
645
+ 566,
646
+ 575
647
+ ],
648
+ [
649
+ 580,
650
+ 598
651
+ ],
652
+ [
653
+ 601,
654
+ 606
655
+ ],
656
+ [
657
+ 634,
658
+ 640
659
+ ],
660
+ [
661
+ 646,
662
+ 653
663
+ ],
664
+ [
665
+ 658,
666
+ 674
667
+ ],
668
+ [
669
+ 717,
670
+ 722
671
+ ],
672
+ [
673
+ 738,
674
+ 744
675
+ ],
676
+ [
677
+ 749,
678
+ 772
679
+ ],
680
+ [
681
+ 790,
682
+ 795
683
+ ]
684
+ ],
685
+ "right_ranges": [
686
+ [
687
+ 33,
688
+ 42
689
+ ],
690
+ [
691
+ 175,
692
+ 182
693
+ ],
694
+ [
695
+ 251,
696
+ 258
697
+ ],
698
+ [
699
+ 399,
700
+ 420
701
+ ],
702
+ [
703
+ 426,
704
+ 435
705
+ ],
706
+ [
707
+ 460,
708
+ 467
709
+ ],
710
+ [
711
+ 472,
712
+ 489
713
+ ],
714
+ [
715
+ 519,
716
+ 540
717
+ ],
718
+ [
719
+ 554,
720
+ 560
721
+ ],
722
+ [
723
+ 566,
724
+ 575
725
+ ],
726
+ [
727
+ 580,
728
+ 598
729
+ ],
730
+ [
731
+ 601,
732
+ 612
733
+ ],
734
+ [
735
+ 634,
736
+ 640
737
+ ],
738
+ [
739
+ 646,
740
+ 653
741
+ ],
742
+ [
743
+ 658,
744
+ 663
745
+ ],
746
+ [
747
+ 666,
748
+ 674
749
+ ],
750
+ [
751
+ 717,
752
+ 722
753
+ ],
754
+ [
755
+ 738,
756
+ 744
757
+ ],
758
+ [
759
+ 749,
760
+ 772
761
+ ],
762
+ [
763
+ 790,
764
+ 795
765
+ ]
766
+ ]
767
+ }
768
+ ]
769
+ },
770
+ {
771
+ "left_pr_number": 44072,
772
+ "right_pr_number": 44085,
773
+ "code_similarity": 0.137,
774
+ "size_similarity": 0.387,
775
+ "file_overlap": 0.0,
776
+ "area_overlap": 0.0,
777
+ "patch_similarity": 0.398,
778
+ "shared_filenames": [],
779
+ "shared_file_areas": []
780
+ },
781
+ {
782
+ "left_pr_number": 44072,
783
+ "right_pr_number": 44129,
784
+ "code_similarity": 0.074,
785
+ "size_similarity": 0.195,
786
+ "file_overlap": 0.0,
787
+ "area_overlap": 0.0,
788
+ "patch_similarity": 0.231,
789
+ "shared_filenames": [],
790
+ "shared_file_areas": []
791
+ },
792
+ {
793
+ "left_pr_number": 44072,
794
+ "right_pr_number": 44154,
795
+ "code_similarity": 0.175,
796
+ "size_similarity": 0.566,
797
+ "file_overlap": 0.0,
798
+ "area_overlap": 0.0,
799
+ "patch_similarity": 0.414,
800
+ "shared_filenames": [],
801
+ "shared_file_areas": []
802
+ },
803
+ {
804
+ "left_pr_number": 44072,
805
+ "right_pr_number": 44722,
806
+ "code_similarity": 0.124,
807
+ "size_similarity": 0.357,
808
+ "file_overlap": 0.0,
809
+ "area_overlap": 0.0,
810
+ "patch_similarity": 0.347,
811
+ "shared_filenames": [],
812
+ "shared_file_areas": []
813
+ },
814
+ {
815
+ "left_pr_number": 44085,
816
+ "right_pr_number": 44129,
817
+ "code_similarity": 0.141,
818
+ "size_similarity": 0.503,
819
+ "file_overlap": 0.0,
820
+ "area_overlap": 0.0,
821
+ "patch_similarity": 0.272,
822
+ "shared_filenames": [],
823
+ "shared_file_areas": []
824
+ },
825
+ {
826
+ "left_pr_number": 44085,
827
+ "right_pr_number": 44154,
828
+ "code_similarity": 0.18,
829
+ "size_similarity": 0.684,
830
+ "file_overlap": 0.0,
831
+ "area_overlap": 0.0,
832
+ "patch_similarity": 0.289,
833
+ "shared_filenames": [],
834
+ "shared_file_areas": []
835
+ },
836
+ {
837
+ "left_pr_number": 44085,
838
+ "right_pr_number": 44722,
839
+ "code_similarity": 0.728,
840
+ "size_similarity": 0.923,
841
+ "file_overlap": 0.5,
842
+ "area_overlap": 0.791,
843
+ "patch_similarity": 0.78,
844
+ "shared_filenames": [
845
+ "src/transformers/models/gptj/modeling_gptj.py"
846
+ ],
847
+ "shared_file_areas": [
848
+ {
849
+ "filename": "src/transformers/models/gptj/modeling_gptj.py",
850
+ "left_ranges": [
851
+ [
852
+ 33,
853
+ 40
854
+ ],
855
+ [
856
+ 173,
857
+ 180
858
+ ],
859
+ [
860
+ 249,
861
+ 256
862
+ ],
863
+ [
864
+ 397,
865
+ 405
866
+ ],
867
+ [
868
+ 408,
869
+ 419
870
+ ],
871
+ [
872
+ 425,
873
+ 434
874
+ ],
875
+ [
876
+ 459,
877
+ 465
878
+ ],
879
+ [
880
+ 470,
881
+ 489
882
+ ],
883
+ [
884
+ 517,
885
+ 539
886
+ ],
887
+ [
888
+ 553,
889
+ 559
890
+ ],
891
+ [
892
+ 565,
893
+ 574
894
+ ],
895
+ [
896
+ 579,
897
+ 597
898
+ ],
899
+ [
900
+ 600,
901
+ 611
902
+ ],
903
+ [
904
+ 633,
905
+ 639
906
+ ],
907
+ [
908
+ 645,
909
+ 652
910
+ ],
911
+ [
912
+ 657,
913
+ 673
914
+ ],
915
+ [
916
+ 716,
917
+ 728
918
+ ],
919
+ [
920
+ 737,
921
+ 743
922
+ ],
923
+ [
924
+ 748,
925
+ 771
926
+ ],
927
+ [
928
+ 789,
929
+ 794
930
+ ]
931
+ ],
932
+ "right_ranges": [
933
+ [
934
+ 33,
935
+ 42
936
+ ],
937
+ [
938
+ 175,
939
+ 182
940
+ ],
941
+ [
942
+ 251,
943
+ 258
944
+ ],
945
+ [
946
+ 399,
947
+ 420
948
+ ],
949
+ [
950
+ 426,
951
+ 435
952
+ ],
953
+ [
954
+ 460,
955
+ 467
956
+ ],
957
+ [
958
+ 472,
959
+ 489
960
+ ],
961
+ [
962
+ 519,
963
+ 540
964
+ ],
965
+ [
966
+ 554,
967
+ 560
968
+ ],
969
+ [
970
+ 566,
971
+ 575
972
+ ],
973
+ [
974
+ 580,
975
+ 598
976
+ ],
977
+ [
978
+ 601,
979
+ 612
980
+ ],
981
+ [
982
+ 634,
983
+ 640
984
+ ],
985
+ [
986
+ 646,
987
+ 653
988
+ ],
989
+ [
990
+ 658,
991
+ 663
992
+ ],
993
+ [
994
+ 666,
995
+ 674
996
+ ],
997
+ [
998
+ 717,
999
+ 722
1000
+ ],
1001
+ [
1002
+ 738,
1003
+ 744
1004
+ ],
1005
+ [
1006
+ 749,
1007
+ 772
1008
+ ],
1009
+ [
1010
+ 790,
1011
+ 795
1012
+ ]
1013
+ ]
1014
+ }
1015
+ ]
1016
+ },
1017
+ {
1018
+ "left_pr_number": 44129,
1019
+ "right_pr_number": 44154,
1020
+ "code_similarity": 0.099,
1021
+ "size_similarity": 0.344,
1022
+ "file_overlap": 0.0,
1023
+ "area_overlap": 0.0,
1024
+ "patch_similarity": 0.199,
1025
+ "shared_filenames": [],
1026
+ "shared_file_areas": []
1027
+ },
1028
+ {
1029
+ "left_pr_number": 44129,
1030
+ "right_pr_number": 44722,
1031
+ "code_similarity": 0.146,
1032
+ "size_similarity": 0.545,
1033
+ "file_overlap": 0.0,
1034
+ "area_overlap": 0.0,
1035
+ "patch_similarity": 0.247,
1036
+ "shared_filenames": [],
1037
+ "shared_file_areas": []
1038
+ },
1039
+ {
1040
+ "left_pr_number": 44154,
1041
+ "right_pr_number": 44722,
1042
+ "code_similarity": 0.164,
1043
+ "size_similarity": 0.631,
1044
+ "file_overlap": 0.0,
1045
+ "area_overlap": 0.0,
1046
+ "patch_similarity": 0.255,
1047
+ "shared_filenames": [],
1048
+ "shared_file_areas": []
1049
+ }
1050
+ ]
1051
+ },
1052
+ {
1053
+ "cluster_id": "cluster-46752-7",
1054
+ "summary": "Cluster of 1 issues and 6 PRs centered on issue #46752.",
1055
+ "status": "open",
1056
+ "confidence": 0.8,
1057
+ "canonical_issue_number": 46752,
1058
+ "canonical_pr_number": 46793,
1059
+ "issue_numbers": [
1060
+ 46752
1061
+ ],
1062
+ "pr_numbers": [
1063
+ 46753,
1064
+ 46778,
1065
+ 46789,
1066
+ 46791,
1067
+ 46793,
1068
+ 46794
1069
+ ],
1070
+ "evidence_types": [
1071
+ "closing_reference",
1072
+ "shared_issue_target",
1073
+ "soft_similarity"
1074
+ ],
1075
+ "pr_comparisons": [
1076
+ {
1077
+ "left_pr_number": 46753,
1078
+ "right_pr_number": 46778,
1079
+ "code_similarity": 0.825,
1080
+ "size_similarity": 0.882,
1081
+ "file_overlap": 1.0,
1082
+ "area_overlap": 0.846,
1083
+ "patch_similarity": 0.346,
1084
+ "shared_filenames": [
1085
+ "src/transformers/tokenization_utils_base.py",
1086
+ "tests/test_tokenization_common.py"
1087
+ ],
1088
+ "shared_file_areas": [
1089
+ {
1090
+ "filename": "src/transformers/tokenization_utils_base.py",
1091
+ "left_ranges": [
1092
+ [
1093
+ 3086,
1094
+ 3094
1095
+ ]
1096
+ ],
1097
+ "right_ranges": [
1098
+ [
1099
+ 3086,
1100
+ 3096
1101
+ ]
1102
+ ]
1103
+ },
1104
+ {
1105
+ "filename": "tests/test_tokenization_common.py",
1106
+ "left_ranges": [
1107
+ [
1108
+ 1010,
1109
+ 1022
1110
+ ]
1111
+ ],
1112
+ "right_ranges": [
1113
+ [
1114
+ 1010,
1115
+ 1024
1116
+ ]
1117
+ ]
1118
+ }
1119
+ ]
1120
+ },
1121
+ {
1122
+ "left_pr_number": 46753,
1123
+ "right_pr_number": 46789,
1124
+ "code_similarity": 0.598,
1125
+ "size_similarity": 0.533,
1126
+ "file_overlap": 0.5,
1127
+ "area_overlap": 0.9,
1128
+ "patch_similarity": 0.176,
1129
+ "shared_filenames": [
1130
+ "src/transformers/tokenization_utils_base.py"
1131
+ ],
1132
+ "shared_file_areas": [
1133
+ {
1134
+ "filename": "src/transformers/tokenization_utils_base.py",
1135
+ "left_ranges": [
1136
+ [
1137
+ 3086,
1138
+ 3094
1139
+ ]
1140
+ ],
1141
+ "right_ranges": [
1142
+ [
1143
+ 3086,
1144
+ 3095
1145
+ ]
1146
+ ]
1147
+ }
1148
+ ]
1149
+ },
1150
+ {
1151
+ "left_pr_number": 46753,
1152
+ "right_pr_number": 46791,
1153
+ "code_similarity": 0.529,
1154
+ "size_similarity": 0.4,
1155
+ "file_overlap": 0.5,
1156
+ "area_overlap": 0.778,
1157
+ "patch_similarity": 0.182,
1158
+ "shared_filenames": [
1159
+ "src/transformers/tokenization_utils_base.py"
1160
+ ],
1161
+ "shared_file_areas": [
1162
+ {
1163
+ "filename": "src/transformers/tokenization_utils_base.py",
1164
+ "left_ranges": [
1165
+ [
1166
+ 3086,
1167
+ 3094
1168
+ ]
1169
+ ],
1170
+ "right_ranges": [
1171
+ [
1172
+ 3086,
1173
+ 3092
1174
+ ]
1175
+ ]
1176
+ }
1177
+ ]
1178
+ },
1179
+ {
1180
+ "left_pr_number": 46753,
1181
+ "right_pr_number": 46793,
1182
+ "code_similarity": 0.379,
1183
+ "size_similarity": 0.455,
1184
+ "file_overlap": 0.2,
1185
+ "area_overlap": 0.529,
1186
+ "patch_similarity": 0.288,
1187
+ "shared_filenames": [
1188
+ "src/transformers/tokenization_utils_base.py"
1189
+ ],
1190
+ "shared_file_areas": [
1191
+ {
1192
+ "filename": "src/transformers/tokenization_utils_base.py",
1193
+ "left_ranges": [
1194
+ [
1195
+ 3086,
1196
+ 3094
1197
+ ]
1198
+ ],
1199
+ "right_ranges": [
1200
+ [
1201
+ 3086,
1202
+ 3095
1203
+ ],
1204
+ [
1205
+ 3450,
1206
+ 3456
1207
+ ]
1208
+ ]
1209
+ }
1210
+ ]
1211
+ },
1212
+ {
1213
+ "left_pr_number": 46753,
1214
+ "right_pr_number": 46794,
1215
+ "code_similarity": 0.529,
1216
+ "size_similarity": 0.4,
1217
+ "file_overlap": 0.5,
1218
+ "area_overlap": 0.778,
1219
+ "patch_similarity": 0.182,
1220
+ "shared_filenames": [
1221
+ "src/transformers/tokenization_utils_base.py"
1222
+ ],
1223
+ "shared_file_areas": [
1224
+ {
1225
+ "filename": "src/transformers/tokenization_utils_base.py",
1226
+ "left_ranges": [
1227
+ [
1228
+ 3086,
1229
+ 3094
1230
+ ]
1231
+ ],
1232
+ "right_ranges": [
1233
+ [
1234
+ 3086,
1235
+ 3092
1236
+ ]
1237
+ ]
1238
+ }
1239
+ ]
1240
+ },
1241
+ {
1242
+ "left_pr_number": 46778,
1243
+ "right_pr_number": 46789,
1244
+ "code_similarity": 0.586,
1245
+ "size_similarity": 0.471,
1246
+ "file_overlap": 0.5,
1247
+ "area_overlap": 0.909,
1248
+ "patch_similarity": 0.158,
1249
+ "shared_filenames": [
1250
+ "src/transformers/tokenization_utils_base.py"
1251
+ ],
1252
+ "shared_file_areas": [
1253
+ {
1254
+ "filename": "src/transformers/tokenization_utils_base.py",
1255
+ "left_ranges": [
1256
+ [
1257
+ 3086,
1258
+ 3096
1259
+ ]
1260
+ ],
1261
+ "right_ranges": [
1262
+ [
1263
+ 3086,
1264
+ 3095
1265
+ ]
1266
+ ]
1267
+ }
1268
+ ]
1269
+ },
1270
+ {
1271
+ "left_pr_number": 46778,
1272
+ "right_pr_number": 46791,
1273
+ "code_similarity": 0.468,
1274
+ "size_similarity": 0.353,
1275
+ "file_overlap": 0.5,
1276
+ "area_overlap": 0.636,
1277
+ "patch_similarity": 0.162,
1278
+ "shared_filenames": [
1279
+ "src/transformers/tokenization_utils_base.py"
1280
+ ],
1281
+ "shared_file_areas": [
1282
+ {
1283
+ "filename": "src/transformers/tokenization_utils_base.py",
1284
+ "left_ranges": [
1285
+ [
1286
+ 3086,
1287
+ 3096
1288
+ ]
1289
+ ],
1290
+ "right_ranges": [
1291
+ [
1292
+ 3086,
1293
+ 3092
1294
+ ]
1295
+ ]
1296
+ }
1297
+ ]
1298
+ },
1299
+ {
1300
+ "left_pr_number": 46778,
1301
+ "right_pr_number": 46793,
1302
+ "code_similarity": 0.401,
1303
+ "size_similarity": 0.515,
1304
+ "file_overlap": 0.2,
1305
+ "area_overlap": 0.556,
1306
+ "patch_similarity": 0.291,
1307
+ "shared_filenames": [
1308
+ "src/transformers/tokenization_utils_base.py"
1309
+ ],
1310
+ "shared_file_areas": [
1311
+ {
1312
+ "filename": "src/transformers/tokenization_utils_base.py",
1313
+ "left_ranges": [
1314
+ [
1315
+ 3086,
1316
+ 3096
1317
+ ]
1318
+ ],
1319
+ "right_ranges": [
1320
+ [
1321
+ 3086,
1322
+ 3095
1323
+ ],
1324
+ [
1325
+ 3450,
1326
+ 3456
1327
+ ]
1328
+ ]
1329
+ }
1330
+ ]
1331
+ },
1332
+ {
1333
+ "left_pr_number": 46778,
1334
+ "right_pr_number": 46794,
1335
+ "code_similarity": 0.468,
1336
+ "size_similarity": 0.353,
1337
+ "file_overlap": 0.5,
1338
+ "area_overlap": 0.636,
1339
+ "patch_similarity": 0.162,
1340
+ "shared_filenames": [
1341
+ "src/transformers/tokenization_utils_base.py"
1342
+ ],
1343
+ "shared_file_areas": [
1344
+ {
1345
+ "filename": "src/transformers/tokenization_utils_base.py",
1346
+ "left_ranges": [
1347
+ [
1348
+ 3086,
1349
+ 3096
1350
+ ]
1351
+ ],
1352
+ "right_ranges": [
1353
+ [
1354
+ 3086,
1355
+ 3092
1356
+ ]
1357
+ ]
1358
+ }
1359
+ ]
1360
+ },
1361
+ {
1362
+ "left_pr_number": 46789,
1363
+ "right_pr_number": 46791,
1364
+ "code_similarity": 0.789,
1365
+ "size_similarity": 0.75,
1366
+ "file_overlap": 1.0,
1367
+ "area_overlap": 0.7,
1368
+ "patch_similarity": 0.625,
1369
+ "shared_filenames": [
1370
+ "src/transformers/tokenization_utils_base.py"
1371
+ ],
1372
+ "shared_file_areas": [
1373
+ {
1374
+ "filename": "src/transformers/tokenization_utils_base.py",
1375
+ "left_ranges": [
1376
+ [
1377
+ 3086,
1378
+ 3095
1379
+ ]
1380
+ ],
1381
+ "right_ranges": [
1382
+ [
1383
+ 3086,
1384
+ 3092
1385
+ ]
1386
+ ]
1387
+ }
1388
+ ]
1389
+ },
1390
+ {
1391
+ "left_pr_number": 46789,
1392
+ "right_pr_number": 46793,
1393
+ "code_similarity": 0.36,
1394
+ "size_similarity": 0.242,
1395
+ "file_overlap": 0.25,
1396
+ "area_overlap": 0.588,
1397
+ "patch_similarity": 0.206,
1398
+ "shared_filenames": [
1399
+ "src/transformers/tokenization_utils_base.py"
1400
+ ],
1401
+ "shared_file_areas": [
1402
+ {
1403
+ "filename": "src/transformers/tokenization_utils_base.py",
1404
+ "left_ranges": [
1405
+ [
1406
+ 3086,
1407
+ 3095
1408
+ ]
1409
+ ],
1410
+ "right_ranges": [
1411
+ [
1412
+ 3086,
1413
+ 3095
1414
+ ],
1415
+ [
1416
+ 3450,
1417
+ 3456
1418
+ ]
1419
+ ]
1420
+ }
1421
+ ]
1422
+ },
1423
+ {
1424
+ "left_pr_number": 46789,
1425
+ "right_pr_number": 46794,
1426
+ "code_similarity": 0.789,
1427
+ "size_similarity": 0.75,
1428
+ "file_overlap": 1.0,
1429
+ "area_overlap": 0.7,
1430
+ "patch_similarity": 0.625,
1431
+ "shared_filenames": [
1432
+ "src/transformers/tokenization_utils_base.py"
1433
+ ],
1434
+ "shared_file_areas": [
1435
+ {
1436
+ "filename": "src/transformers/tokenization_utils_base.py",
1437
+ "left_ranges": [
1438
+ [
1439
+ 3086,
1440
+ 3095
1441
+ ]
1442
+ ],
1443
+ "right_ranges": [
1444
+ [
1445
+ 3086,
1446
+ 3092
1447
+ ]
1448
+ ]
1449
+ }
1450
+ ]
1451
+ },
1452
+ {
1453
+ "left_pr_number": 46791,
1454
+ "right_pr_number": 46793,
1455
+ "code_similarity": 0.282,
1456
+ "size_similarity": 0.182,
1457
+ "file_overlap": 0.25,
1458
+ "area_overlap": 0.412,
1459
+ "patch_similarity": 0.176,
1460
+ "shared_filenames": [
1461
+ "src/transformers/tokenization_utils_base.py"
1462
+ ],
1463
+ "shared_file_areas": [
1464
+ {
1465
+ "filename": "src/transformers/tokenization_utils_base.py",
1466
+ "left_ranges": [
1467
+ [
1468
+ 3086,
1469
+ 3092
1470
+ ]
1471
+ ],
1472
+ "right_ranges": [
1473
+ [
1474
+ 3086,
1475
+ 3095
1476
+ ],
1477
+ [
1478
+ 3450,
1479
+ 3456
1480
+ ]
1481
+ ]
1482
+ }
1483
+ ]
1484
+ },
1485
+ {
1486
+ "left_pr_number": 46791,
1487
+ "right_pr_number": 46794,
1488
+ "code_similarity": 1.0,
1489
+ "size_similarity": 1.0,
1490
+ "file_overlap": 1.0,
1491
+ "area_overlap": 1.0,
1492
+ "patch_similarity": 1.0,
1493
+ "shared_filenames": [
1494
+ "src/transformers/tokenization_utils_base.py"
1495
+ ],
1496
+ "shared_file_areas": [
1497
+ {
1498
+ "filename": "src/transformers/tokenization_utils_base.py",
1499
+ "left_ranges": [
1500
+ [
1501
+ 3086,
1502
+ 3092
1503
+ ]
1504
+ ],
1505
+ "right_ranges": [
1506
+ [
1507
+ 3086,
1508
+ 3092
1509
+ ]
1510
+ ]
1511
+ }
1512
+ ]
1513
+ },
1514
+ {
1515
+ "left_pr_number": 46793,
1516
+ "right_pr_number": 46794,
1517
+ "code_similarity": 0.282,
1518
+ "size_similarity": 0.182,
1519
+ "file_overlap": 0.25,
1520
+ "area_overlap": 0.412,
1521
+ "patch_similarity": 0.176,
1522
+ "shared_filenames": [
1523
+ "src/transformers/tokenization_utils_base.py"
1524
+ ],
1525
+ "shared_file_areas": [
1526
+ {
1527
+ "filename": "src/transformers/tokenization_utils_base.py",
1528
+ "left_ranges": [
1529
+ [
1530
+ 3086,
1531
+ 3095
1532
+ ],
1533
+ [
1534
+ 3450,
1535
+ 3456
1536
+ ]
1537
+ ],
1538
+ "right_ranges": [
1539
+ [
1540
+ 3086,
1541
+ 3092
1542
+ ]
1543
+ ]
1544
+ }
1545
+ ]
1546
+ }
1547
+ ]
1548
+ },
1549
+ {
1550
+ "cluster_id": "cluster-41211-3",
1551
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #41211.",
1552
+ "status": "open",
1553
+ "confidence": 0.55,
1554
+ "canonical_issue_number": 41211,
1555
+ "canonical_pr_number": 44339,
1556
+ "issue_numbers": [
1557
+ 41211
1558
+ ],
1559
+ "pr_numbers": [
1560
+ 41356,
1561
+ 44339
1562
+ ],
1563
+ "evidence_types": [
1564
+ "closing_reference"
1565
+ ],
1566
+ "pr_comparisons": [
1567
+ {
1568
+ "left_pr_number": 41356,
1569
+ "right_pr_number": 44339,
1570
+ "code_similarity": 0.155,
1571
+ "size_similarity": 0.094,
1572
+ "file_overlap": 0.273,
1573
+ "area_overlap": 0.078,
1574
+ "patch_similarity": 0.183,
1575
+ "shared_filenames": [
1576
+ "docs/source/en/model_doc/deimv2.md",
1577
+ "src/transformers/models/deimv2/__init__.py",
1578
+ "src/transformers/models/deimv2/configuration_deimv2.py",
1579
+ "src/transformers/models/deimv2/modeling_deimv2.py",
1580
+ "tests/models/deimv2/__init__.py",
1581
+ "tests/models/deimv2/test_modeling_deimv2.py"
1582
+ ],
1583
+ "shared_file_areas": [
1584
+ {
1585
+ "filename": "docs/source/en/model_doc/deimv2.md",
1586
+ "left_ranges": [
1587
+ [
1588
+ 1,
1589
+ 132
1590
+ ]
1591
+ ],
1592
+ "right_ranges": [
1593
+ [
1594
+ 1,
1595
+ 65
1596
+ ]
1597
+ ]
1598
+ },
1599
+ {
1600
+ "filename": "src/transformers/models/deimv2/__init__.py",
1601
+ "left_ranges": [
1602
+ [
1603
+ 1,
1604
+ 15
1605
+ ]
1606
+ ],
1607
+ "right_ranges": [
1608
+ [
1609
+ 1,
1610
+ 29
1611
+ ]
1612
+ ]
1613
+ },
1614
+ {
1615
+ "filename": "src/transformers/models/deimv2/configuration_deimv2.py",
1616
+ "left_ranges": [
1617
+ [
1618
+ 1,
1619
+ 74
1620
+ ]
1621
+ ],
1622
+ "right_ranges": [
1623
+ [
1624
+ 1,
1625
+ 266
1626
+ ]
1627
+ ]
1628
+ },
1629
+ {
1630
+ "filename": "tests/models/deimv2/test_modeling_deimv2.py",
1631
+ "left_ranges": [
1632
+ [
1633
+ 1,
1634
+ 15
1635
+ ]
1636
+ ],
1637
+ "right_ranges": [
1638
+ [
1639
+ 1,
1640
+ 1734
1641
+ ]
1642
+ ]
1643
+ }
1644
+ ]
1645
+ }
1646
+ ]
1647
+ },
1648
+ {
1649
+ "cluster_id": "cluster-41115-3",
1650
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #41115.",
1651
+ "status": "open",
1652
+ "confidence": 0.55,
1653
+ "canonical_issue_number": 41115,
1654
+ "canonical_pr_number": 41116,
1655
+ "issue_numbers": [
1656
+ 41115
1657
+ ],
1658
+ "pr_numbers": [
1659
+ 41116,
1660
+ 45613
1661
+ ],
1662
+ "evidence_types": [
1663
+ "closing_reference"
1664
+ ],
1665
+ "pr_comparisons": [
1666
+ {
1667
+ "left_pr_number": 41116,
1668
+ "right_pr_number": 45613,
1669
+ "code_similarity": 0.811,
1670
+ "size_similarity": 0.9,
1671
+ "file_overlap": 0.75,
1672
+ "area_overlap": 0.882,
1673
+ "patch_similarity": 0.648,
1674
+ "shared_filenames": [
1675
+ "docs/source/en/_toctree.yml",
1676
+ "docs/source/en/model_doc/minicpm3.md",
1677
+ "src/transformers/models/auto/modeling_auto.py",
1678
+ "src/transformers/models/minicpm3/__init__.py",
1679
+ "src/transformers/models/minicpm3/configuration_minicpm3.py",
1680
+ "src/transformers/models/minicpm3/modeling_minicpm3.py",
1681
+ "src/transformers/models/minicpm3/modular_minicpm3.py",
1682
+ "tests/models/minicpm3/__init__.py",
1683
+ "tests/models/minicpm3/test_modeling_minicpm3.py"
1684
+ ],
1685
+ "shared_file_areas": [
1686
+ {
1687
+ "filename": "docs/source/en/_toctree.yml",
1688
+ "left_ranges": [
1689
+ [
1690
+ 737,
1691
+ 744
1692
+ ]
1693
+ ],
1694
+ "right_ranges": [
1695
+ [
1696
+ 730,
1697
+ 737
1698
+ ]
1699
+ ]
1700
+ },
1701
+ {
1702
+ "filename": "docs/source/en/model_doc/minicpm3.md",
1703
+ "left_ranges": [
1704
+ [
1705
+ 1,
1706
+ 67
1707
+ ]
1708
+ ],
1709
+ "right_ranges": [
1710
+ [
1711
+ 1,
1712
+ 45
1713
+ ]
1714
+ ]
1715
+ },
1716
+ {
1717
+ "filename": "src/transformers/models/auto/modeling_auto.py",
1718
+ "left_ranges": [
1719
+ [
1720
+ 301,
1721
+ 307
1722
+ ],
1723
+ [
1724
+ 742,
1725
+ 748
1726
+ ],
1727
+ [
1728
+ 1360,
1729
+ 1366
1730
+ ]
1731
+ ],
1732
+ "right_ranges": [
1733
+ [
1734
+ 720,
1735
+ 726
1736
+ ],
1737
+ [
1738
+ 1331,
1739
+ 1337
1740
+ ]
1741
+ ]
1742
+ },
1743
+ {
1744
+ "filename": "src/transformers/models/minicpm3/__init__.py",
1745
+ "left_ranges": [
1746
+ [
1747
+ 1,
1748
+ 29
1749
+ ]
1750
+ ],
1751
+ "right_ranges": [
1752
+ [
1753
+ 1,
1754
+ 29
1755
+ ]
1756
+ ]
1757
+ },
1758
+ {
1759
+ "filename": "src/transformers/models/minicpm3/configuration_minicpm3.py",
1760
+ "left_ranges": [
1761
+ [
1762
+ 1,
1763
+ 150
1764
+ ]
1765
+ ],
1766
+ "right_ranges": [
1767
+ [
1768
+ 1,
1769
+ 126
1770
+ ]
1771
+ ]
1772
+ },
1773
+ {
1774
+ "filename": "src/transformers/models/minicpm3/modeling_minicpm3.py",
1775
+ "left_ranges": [
1776
+ [
1777
+ 1,
1778
+ 575
1779
+ ]
1780
+ ],
1781
+ "right_ranges": [
1782
+ [
1783
+ 1,
1784
+ 522
1785
+ ]
1786
+ ]
1787
+ },
1788
+ {
1789
+ "filename": "src/transformers/models/minicpm3/modular_minicpm3.py",
1790
+ "left_ranges": [
1791
+ [
1792
+ 1,
1793
+ 354
1794
+ ]
1795
+ ],
1796
+ "right_ranges": [
1797
+ [
1798
+ 1,
1799
+ 342
1800
+ ]
1801
+ ]
1802
+ },
1803
+ {
1804
+ "filename": "tests/models/minicpm3/test_modeling_minicpm3.py",
1805
+ "left_ranges": [
1806
+ [
1807
+ 1,
1808
+ 135
1809
+ ]
1810
+ ],
1811
+ "right_ranges": [
1812
+ [
1813
+ 1,
1814
+ 136
1815
+ ]
1816
+ ]
1817
+ }
1818
+ ]
1819
+ }
1820
+ ]
1821
+ },
1822
+ {
1823
+ "cluster_id": "cluster-43240-3",
1824
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.",
1825
+ "status": "open",
1826
+ "confidence": 0.75,
1827
+ "canonical_issue_number": 43240,
1828
+ "canonical_pr_number": 43251,
1829
+ "issue_numbers": [
1830
+ 43240
1831
+ ],
1832
+ "pr_numbers": [
1833
+ 43251,
1834
+ 43254
1835
+ ],
1836
+ "evidence_types": [
1837
+ "closing_reference",
1838
+ "shared_issue_target"
1839
+ ],
1840
+ "pr_comparisons": [
1841
+ {
1842
+ "left_pr_number": 43251,
1843
+ "right_pr_number": 43254,
1844
+ "code_similarity": 0.794,
1845
+ "size_similarity": 0.64,
1846
+ "file_overlap": 1.0,
1847
+ "area_overlap": 0.667,
1848
+ "patch_similarity": 0.882,
1849
+ "shared_filenames": [
1850
+ "src/transformers/loss/loss_utils.py"
1851
+ ],
1852
+ "shared_file_areas": [
1853
+ {
1854
+ "filename": "src/transformers/loss/loss_utils.py",
1855
+ "left_ranges": [
1856
+ [
1857
+ 30,
1858
+ 50
1859
+ ]
1860
+ ],
1861
+ "right_ranges": [
1862
+ [
1863
+ 30,
1864
+ 43
1865
+ ]
1866
+ ]
1867
+ }
1868
+ ]
1869
+ }
1870
+ ]
1871
+ },
1872
+ {
1873
+ "cluster_id": "cluster-45647-3",
1874
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #45647.",
1875
+ "status": "open",
1876
+ "confidence": 0.75,
1877
+ "canonical_issue_number": 45647,
1878
+ "canonical_pr_number": 45738,
1879
+ "issue_numbers": [
1880
+ 45647
1881
+ ],
1882
+ "pr_numbers": [
1883
+ 45737,
1884
+ 45738
1885
+ ],
1886
+ "evidence_types": [
1887
+ "closing_reference",
1888
+ "shared_issue_target"
1889
+ ],
1890
+ "pr_comparisons": [
1891
+ {
1892
+ "left_pr_number": 45737,
1893
+ "right_pr_number": 45738,
1894
+ "code_similarity": 0.526,
1895
+ "size_similarity": 0.11,
1896
+ "file_overlap": 0.5,
1897
+ "area_overlap": 1.0,
1898
+ "patch_similarity": 0.024,
1899
+ "shared_filenames": [
1900
+ "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py"
1901
+ ],
1902
+ "shared_file_areas": [
1903
+ {
1904
+ "filename": "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py",
1905
+ "left_ranges": [
1906
+ [
1907
+ 491,
1908
+ 497
1909
+ ]
1910
+ ],
1911
+ "right_ranges": [
1912
+ [
1913
+ 491,
1914
+ 497
1915
+ ]
1916
+ ]
1917
+ }
1918
+ ]
1919
+ }
1920
+ ]
1921
+ },
1922
+ {
1923
+ "cluster_id": "cluster-45987-3",
1924
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #45987.",
1925
+ "status": "open",
1926
+ "confidence": 0.75,
1927
+ "canonical_issue_number": 45987,
1928
+ "canonical_pr_number": 46173,
1929
+ "issue_numbers": [
1930
+ 45987
1931
+ ],
1932
+ "pr_numbers": [
1933
+ 46173,
1934
+ 46474
1935
+ ],
1936
+ "evidence_types": [
1937
+ "closing_reference",
1938
+ "shared_issue_target"
1939
+ ],
1940
+ "pr_comparisons": [
1941
+ {
1942
+ "left_pr_number": 46173,
1943
+ "right_pr_number": 46474,
1944
+ "code_similarity": 0.65,
1945
+ "size_similarity": 1.0,
1946
+ "file_overlap": 1.0,
1947
+ "area_overlap": 0.0,
1948
+ "patch_similarity": 1.0,
1949
+ "shared_filenames": [
1950
+ "src/transformers/cache_utils.py"
1951
+ ],
1952
+ "shared_file_areas": [
1953
+ {
1954
+ "filename": "src/transformers/cache_utils.py",
1955
+ "left_ranges": [
1956
+ [
1957
+ 69,
1958
+ 75
1959
+ ],
1960
+ [
1961
+ 375,
1962
+ 381
1963
+ ]
1964
+ ],
1965
+ "right_ranges": [
1966
+ [
1967
+ 79,
1968
+ 85
1969
+ ],
1970
+ [
1971
+ 461,
1972
+ 467
1973
+ ]
1974
+ ]
1975
+ }
1976
+ ]
1977
+ }
1978
+ ]
1979
+ },
1980
+ {
1981
+ "cluster_id": "cluster-46159-3",
1982
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #46159.",
1983
+ "status": "open",
1984
+ "confidence": 0.55,
1985
+ "canonical_issue_number": 46159,
1986
+ "canonical_pr_number": 46174,
1987
+ "issue_numbers": [
1988
+ 46159
1989
+ ],
1990
+ "pr_numbers": [
1991
+ 46174,
1992
+ 46780
1993
+ ],
1994
+ "evidence_types": [
1995
+ "closing_reference"
1996
+ ],
1997
+ "pr_comparisons": [
1998
+ {
1999
+ "left_pr_number": 46174,
2000
+ "right_pr_number": 46780,
2001
+ "code_similarity": 0.841,
2002
+ "size_similarity": 0.87,
2003
+ "file_overlap": 1.0,
2004
+ "area_overlap": 0.767,
2005
+ "patch_similarity": 0.657,
2006
+ "shared_filenames": [
2007
+ "src/transformers/models/aimv2/modeling_aimv2.py",
2008
+ "src/transformers/models/d_fine/modeling_d_fine.py",
2009
+ "src/transformers/models/deimv2/modeling_deimv2.py",
2010
+ "src/transformers/models/pp_doclayout_v2/modeling_pp_doclayout_v2.py",
2011
+ "src/transformers/models/pp_doclayout_v3/modeling_pp_doclayout_v3.py",
2012
+ "src/transformers/models/rt_detr/modeling_rt_detr.py",
2013
+ "src/transformers/models/rt_detr_v2/modeling_rt_detr_v2.py",
2014
+ "src/transformers/models/vit_mae/modeling_vit_mae.py",
2015
+ "src/transformers/models/vit_mae/modular_vit_mae.py"
2016
+ ],
2017
+ "shared_file_areas": [
2018
+ {
2019
+ "filename": "src/transformers/models/aimv2/modeling_aimv2.py",
2020
+ "left_ranges": [
2021
+ [
2022
+ 134,
2023
+ 141
2024
+ ],
2025
+ [
2026
+ 144,
2027
+ 154
2028
+ ],
2029
+ [
2030
+ 157,
2031
+ 165
2032
+ ]
2033
+ ],
2034
+ "right_ranges": [
2035
+ [
2036
+ 134,
2037
+ 140
2038
+ ],
2039
+ [
2040
+ 143,
2041
+ 153
2042
+ ],
2043
+ [
2044
+ 156,
2045
+ 162
2046
+ ]
2047
+ ]
2048
+ },
2049
+ {
2050
+ "filename": "src/transformers/models/d_fine/modeling_d_fine.py",
2051
+ "left_ranges": [
2052
+ [
2053
+ 631,
2054
+ 638
2055
+ ],
2056
+ [
2057
+ 641,
2058
+ 651
2059
+ ],
2060
+ [
2061
+ 654,
2062
+ 662
2063
+ ]
2064
+ ],
2065
+ "right_ranges": [
2066
+ [
2067
+ 631,
2068
+ 637
2069
+ ],
2070
+ [
2071
+ 640,
2072
+ 650
2073
+ ],
2074
+ [
2075
+ 653,
2076
+ 659
2077
+ ]
2078
+ ]
2079
+ },
2080
+ {
2081
+ "filename": "src/transformers/models/deimv2/modeling_deimv2.py",
2082
+ "left_ranges": [
2083
+ [
2084
+ 703,
2085
+ 710
2086
+ ],
2087
+ [
2088
+ 713,
2089
+ 723
2090
+ ],
2091
+ [
2092
+ 726,
2093
+ 734
2094
+ ]
2095
+ ],
2096
+ "right_ranges": [
2097
+ [
2098
+ 703,
2099
+ 709
2100
+ ],
2101
+ [
2102
+ 712,
2103
+ 722
2104
+ ],
2105
+ [
2106
+ 725,
2107
+ 731
2108
+ ]
2109
+ ]
2110
+ },
2111
+ {
2112
+ "filename": "src/transformers/models/pp_doclayout_v2/modeling_pp_doclayout_v2.py",
2113
+ "left_ranges": [
2114
+ [
2115
+ 1551,
2116
+ 1558
2117
+ ],
2118
+ [
2119
+ 1561,
2120
+ 1571
2121
+ ],
2122
+ [
2123
+ 1574,
2124
+ 1582
2125
+ ]
2126
+ ],
2127
+ "right_ranges": [
2128
+ [
2129
+ 1551,
2130
+ 1557
2131
+ ],
2132
+ [
2133
+ 1560,
2134
+ 1570
2135
+ ],
2136
+ [
2137
+ 1573,
2138
+ 1579
2139
+ ]
2140
+ ]
2141
+ },
2142
+ {
2143
+ "filename": "src/transformers/models/pp_doclayout_v3/modeling_pp_doclayout_v3.py",
2144
+ "left_ranges": [
2145
+ [
2146
+ 804,
2147
+ 811
2148
+ ],
2149
+ [
2150
+ 814,
2151
+ 824
2152
+ ],
2153
+ [
2154
+ 827,
2155
+ 835
2156
+ ]
2157
+ ],
2158
+ "right_ranges": [
2159
+ [
2160
+ 804,
2161
+ 810
2162
+ ],
2163
+ [
2164
+ 813,
2165
+ 823
2166
+ ],
2167
+ [
2168
+ 826,
2169
+ 832
2170
+ ]
2171
+ ]
2172
+ },
2173
+ {
2174
+ "filename": "src/transformers/models/rt_detr/modeling_rt_detr.py",
2175
+ "left_ranges": [
2176
+ [
2177
+ 847,
2178
+ 854
2179
+ ],
2180
+ [
2181
+ 857,
2182
+ 867
2183
+ ],
2184
+ [
2185
+ 870,
2186
+ 878
2187
+ ]
2188
+ ],
2189
+ "right_ranges": [
2190
+ [
2191
+ 847,
2192
+ 853
2193
+ ],
2194
+ [
2195
+ 856,
2196
+ 866
2197
+ ],
2198
+ [
2199
+ 869,
2200
+ 875
2201
+ ]
2202
+ ]
2203
+ },
2204
+ {
2205
+ "filename": "src/transformers/models/rt_detr_v2/modeling_rt_detr_v2.py",
2206
+ "left_ranges": [
2207
+ [
2208
+ 976,
2209
+ 983
2210
+ ],
2211
+ [
2212
+ 986,
2213
+ 996
2214
+ ],
2215
+ [
2216
+ 999,
2217
+ 1007
2218
+ ]
2219
+ ],
2220
+ "right_ranges": [
2221
+ [
2222
+ 976,
2223
+ 982
2224
+ ],
2225
+ [
2226
+ 985,
2227
+ 995
2228
+ ],
2229
+ [
2230
+ 998,
2231
+ 1004
2232
+ ]
2233
+ ]
2234
+ },
2235
+ {
2236
+ "filename": "src/transformers/models/vit_mae/modeling_vit_mae.py",
2237
+ "left_ranges": [
2238
+ [
2239
+ 152,
2240
+ 159
2241
+ ],
2242
+ [
2243
+ 162,
2244
+ 172
2245
+ ],
2246
+ [
2247
+ 175,
2248
+ 183
2249
+ ]
2250
+ ],
2251
+ "right_ranges": [
2252
+ [
2253
+ 152,
2254
+ 158
2255
+ ],
2256
+ [
2257
+ 161,
2258
+ 171
2259
+ ],
2260
+ [
2261
+ 174,
2262
+ 180
2263
+ ]
2264
+ ]
2265
+ },
2266
+ {
2267
+ "filename": "src/transformers/models/vit_mae/modular_vit_mae.py",
2268
+ "left_ranges": [
2269
+ [
2270
+ 62,
2271
+ 69
2272
+ ],
2273
+ [
2274
+ 72,
2275
+ 82
2276
+ ],
2277
+ [
2278
+ 85,
2279
+ 93
2280
+ ]
2281
+ ],
2282
+ "right_ranges": [
2283
+ [
2284
+ 62,
2285
+ 68
2286
+ ],
2287
+ [
2288
+ 71,
2289
+ 81
2290
+ ],
2291
+ [
2292
+ 84,
2293
+ 90
2294
+ ]
2295
+ ]
2296
+ }
2297
+ ]
2298
+ }
2299
+ ]
2300
+ },
2301
+ {
2302
+ "cluster_id": "cluster-45561-3",
2303
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #45561.",
2304
+ "status": "open",
2305
+ "confidence": 0.8,
2306
+ "canonical_issue_number": 45561,
2307
+ "canonical_pr_number": 45645,
2308
+ "issue_numbers": [
2309
+ 45561
2310
+ ],
2311
+ "pr_numbers": [
2312
+ 45639,
2313
+ 45645
2314
+ ],
2315
+ "evidence_types": [
2316
+ "closing_reference",
2317
+ "shared_issue_target",
2318
+ "soft_similarity"
2319
+ ],
2320
+ "pr_comparisons": [
2321
+ {
2322
+ "left_pr_number": 45639,
2323
+ "right_pr_number": 45645,
2324
+ "code_similarity": 0.494,
2325
+ "size_similarity": 0.619,
2326
+ "file_overlap": 0.5,
2327
+ "area_overlap": 0.434,
2328
+ "patch_similarity": 0.451,
2329
+ "shared_filenames": [
2330
+ "src/transformers/testing_utils.py",
2331
+ "tests/utils/test_testing_utils.py"
2332
+ ],
2333
+ "shared_file_areas": [
2334
+ {
2335
+ "filename": "src/transformers/testing_utils.py",
2336
+ "left_ranges": [
2337
+ [
2338
+ 3595,
2339
+ 3602
2340
+ ],
2341
+ [
2342
+ 3819,
2343
+ 3853
2344
+ ]
2345
+ ],
2346
+ "right_ranges": [
2347
+ [
2348
+ 3525,
2349
+ 3558
2350
+ ],
2351
+ [
2352
+ 3782,
2353
+ 3788
2354
+ ]
2355
+ ]
2356
+ },
2357
+ {
2358
+ "filename": "tests/utils/test_testing_utils.py",
2359
+ "left_ranges": [
2360
+ [
2361
+ 1,
2362
+ 86
2363
+ ]
2364
+ ],
2365
+ "right_ranges": [
2366
+ [
2367
+ 1,
2368
+ 114
2369
+ ]
2370
+ ]
2371
+ }
2372
+ ]
2373
+ }
2374
+ ]
2375
+ },
2376
+ {
2377
+ "cluster_id": "cluster-46305-3",
2378
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #46305.",
2379
+ "status": "open",
2380
+ "confidence": 0.8,
2381
+ "canonical_issue_number": 46305,
2382
+ "canonical_pr_number": 46546,
2383
+ "issue_numbers": [
2384
+ 46305
2385
+ ],
2386
+ "pr_numbers": [
2387
+ 46310,
2388
+ 46546
2389
+ ],
2390
+ "evidence_types": [
2391
+ "closing_reference",
2392
+ "shared_issue_target",
2393
+ "soft_similarity"
2394
+ ],
2395
+ "pr_comparisons": [
2396
+ {
2397
+ "left_pr_number": 46310,
2398
+ "right_pr_number": 46546,
2399
+ "code_similarity": 0.566,
2400
+ "size_similarity": 0.444,
2401
+ "file_overlap": 0.5,
2402
+ "area_overlap": 0.867,
2403
+ "patch_similarity": 0.16,
2404
+ "shared_filenames": [
2405
+ "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py",
2406
+ "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py"
2407
+ ],
2408
+ "shared_file_areas": [
2409
+ {
2410
+ "filename": "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py",
2411
+ "left_ranges": [
2412
+ [
2413
+ 356,
2414
+ 362
2415
+ ]
2416
+ ],
2417
+ "right_ranges": [
2418
+ [
2419
+ 355,
2420
+ 361
2421
+ ]
2422
+ ]
2423
+ },
2424
+ {
2425
+ "filename": "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py",
2426
+ "left_ranges": [
2427
+ [
2428
+ 269,
2429
+ 275
2430
+ ]
2431
+ ],
2432
+ "right_ranges": [
2433
+ [
2434
+ 269,
2435
+ 275
2436
+ ]
2437
+ ]
2438
+ }
2439
+ ]
2440
+ }
2441
+ ]
2442
+ },
2443
+ {
2444
+ "cluster_id": "cluster-45081-3",
2445
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #45081.",
2446
+ "status": "open",
2447
+ "confidence": 0.8,
2448
+ "canonical_issue_number": 45081,
2449
+ "canonical_pr_number": 45317,
2450
+ "issue_numbers": [
2451
+ 45081
2452
+ ],
2453
+ "pr_numbers": [
2454
+ 45086,
2455
+ 45317
2456
+ ],
2457
+ "evidence_types": [
2458
+ "closing_reference",
2459
+ "shared_issue_target",
2460
+ "soft_similarity"
2461
+ ],
2462
+ "pr_comparisons": [
2463
+ {
2464
+ "left_pr_number": 45086,
2465
+ "right_pr_number": 45317,
2466
+ "code_similarity": 0.589,
2467
+ "size_similarity": 0.34,
2468
+ "file_overlap": 0.5,
2469
+ "area_overlap": 1.0,
2470
+ "patch_similarity": 0.136,
2471
+ "shared_filenames": [
2472
+ "src/transformers/tokenization_utils_tokenizers.py"
2473
+ ],
2474
+ "shared_file_areas": [
2475
+ {
2476
+ "filename": "src/transformers/tokenization_utils_tokenizers.py",
2477
+ "left_ranges": [
2478
+ [
2479
+ 1360,
2480
+ 1370
2481
+ ],
2482
+ [
2483
+ 1374,
2484
+ 1380
2485
+ ]
2486
+ ],
2487
+ "right_ranges": [
2488
+ [
2489
+ 1360,
2490
+ 1370
2491
+ ],
2492
+ [
2493
+ 1374,
2494
+ 1380
2495
+ ]
2496
+ ]
2497
+ }
2498
+ ]
2499
+ }
2500
+ ]
2501
+ },
2502
+ {
2503
+ "cluster_id": "cluster-46682-3",
2504
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #46682.",
2505
+ "status": "open",
2506
+ "confidence": 0.75,
2507
+ "canonical_issue_number": 46682,
2508
+ "canonical_pr_number": 46788,
2509
+ "issue_numbers": [
2510
+ 46682
2511
+ ],
2512
+ "pr_numbers": [
2513
+ 46757,
2514
+ 46788
2515
+ ],
2516
+ "evidence_types": [
2517
+ "closing_reference",
2518
+ "shared_issue_target"
2519
+ ],
2520
+ "pr_comparisons": [
2521
+ {
2522
+ "left_pr_number": 46757,
2523
+ "right_pr_number": 46788,
2524
+ "code_similarity": 0.263,
2525
+ "size_similarity": 0.464,
2526
+ "file_overlap": 0.5,
2527
+ "area_overlap": 0.0,
2528
+ "patch_similarity": 0.132,
2529
+ "shared_filenames": [
2530
+ "src/transformers/masking_utils.py"
2531
+ ],
2532
+ "shared_file_areas": [
2533
+ {
2534
+ "filename": "src/transformers/masking_utils.py",
2535
+ "left_ranges": [
2536
+ [
2537
+ 856,
2538
+ 862
2539
+ ]
2540
+ ],
2541
+ "right_ranges": [
2542
+ [
2543
+ 709,
2544
+ 717
2545
+ ]
2546
+ ]
2547
+ }
2548
+ ]
2549
+ }
2550
+ ]
2551
+ },
2552
+ {
2553
+ "cluster_id": "cluster-43698-3",
2554
+ "summary": "Cluster of 1 issues and 2 PRs centered on issue #43698.",
2555
+ "status": "open",
2556
+ "confidence": 0.75,
2557
+ "canonical_issue_number": 43698,
2558
+ "canonical_pr_number": 43779,
2559
+ "issue_numbers": [
2560
+ 43698
2561
+ ],
2562
+ "pr_numbers": [
2563
+ 43779,
2564
+ 43816
2565
+ ],
2566
+ "evidence_types": [
2567
+ "closing_reference",
2568
+ "shared_issue_target"
2569
+ ],
2570
+ "pr_comparisons": [
2571
+ {
2572
+ "left_pr_number": 43779,
2573
+ "right_pr_number": 43816,
2574
+ "code_similarity": 0.418,
2575
+ "size_similarity": 0.538,
2576
+ "file_overlap": 1.0,
2577
+ "area_overlap": 0.02,
2578
+ "patch_similarity": 0.02,
2579
+ "shared_filenames": [
2580
+ "src/transformers/integrations/integration_utils.py"
2581
+ ],
2582
+ "shared_file_areas": [
2583
+ {
2584
+ "filename": "src/transformers/integrations/integration_utils.py",
2585
+ "left_ranges": [
2586
+ [
2587
+ 2227,
2588
+ 2241
2589
+ ],
2590
+ [
2591
+ 2303,
2592
+ 2309
2593
+ ]
2594
+ ],
2595
+ "right_ranges": [
2596
+ [
2597
+ 2278,
2598
+ 2291
2599
+ ],
2600
+ [
2601
+ 2309,
2602
+ 2322
2603
+ ]
2604
+ ]
2605
+ }
2606
+ ]
2607
+ }
2608
+ ]
2609
+ },
2610
+ {
2611
+ "cluster_id": "cluster-44018-2",
2612
+ "summary": "Cluster of 2 related pull requests linked by soft_similarity.",
2613
+ "status": "open",
2614
+ "confidence": 0.5,
2615
+ "canonical_issue_number": null,
2616
+ "canonical_pr_number": 44068,
2617
+ "issue_numbers": [],
2618
+ "pr_numbers": [
2619
+ 44018,
2620
+ 44068
2621
+ ],
2622
+ "evidence_types": [
2623
+ "soft_similarity"
2624
+ ],
2625
+ "pr_comparisons": [
2626
+ {
2627
+ "left_pr_number": 44018,
2628
+ "right_pr_number": 44068,
2629
+ "code_similarity": 0.766,
2630
+ "size_similarity": 0.939,
2631
+ "file_overlap": 1.0,
2632
+ "area_overlap": 0.425,
2633
+ "patch_similarity": 0.866,
2634
+ "shared_filenames": [
2635
+ "src/transformers/models/gpt_neo/modeling_gpt_neo.py",
2636
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py"
2637
+ ],
2638
+ "shared_file_areas": [
2639
+ {
2640
+ "filename": "src/transformers/models/gpt_neo/modeling_gpt_neo.py",
2641
+ "left_ranges": [
2642
+ [
2643
+ 26,
2644
+ 46
2645
+ ],
2646
+ [
2647
+ 138,
2648
+ 143
2649
+ ],
2650
+ [
2651
+ 182,
2652
+ 187
2653
+ ],
2654
+ [
2655
+ 283,
2656
+ 295
2657
+ ],
2658
+ [
2659
+ 327,
2660
+ 341
2661
+ ],
2662
+ [
2663
+ 348,
2664
+ 354
2665
+ ],
2666
+ [
2667
+ 360,
2668
+ 369
2669
+ ],
2670
+ [
2671
+ 399,
2672
+ 406
2673
+ ],
2674
+ [
2675
+ 411,
2676
+ 419
2677
+ ],
2678
+ [
2679
+ 428,
2680
+ 433
2681
+ ],
2682
+ [
2683
+ 472,
2684
+ 492
2685
+ ],
2686
+ [
2687
+ 507,
2688
+ 513
2689
+ ],
2690
+ [
2691
+ 519,
2692
+ 528
2693
+ ],
2694
+ [
2695
+ 541,
2696
+ 559
2697
+ ],
2698
+ [
2699
+ 562,
2700
+ 567
2701
+ ],
2702
+ [
2703
+ 595,
2704
+ 601
2705
+ ],
2706
+ [
2707
+ 607,
2708
+ 614
2709
+ ],
2710
+ [
2711
+ 627,
2712
+ 643
2713
+ ],
2714
+ [
2715
+ 685,
2716
+ 690
2717
+ ],
2718
+ [
2719
+ 708,
2720
+ 714
2721
+ ],
2722
+ [
2723
+ 720,
2724
+ 727
2725
+ ],
2726
+ [
2727
+ 740,
2728
+ 757
2729
+ ],
2730
+ [
2731
+ 761,
2732
+ 766
2733
+ ],
2734
+ [
2735
+ 780,
2736
+ 786
2737
+ ],
2738
+ [
2739
+ 791,
2740
+ 798
2741
+ ],
2742
+ [
2743
+ 807,
2744
+ 822
2745
+ ],
2746
+ [
2747
+ 840,
2748
+ 845
2749
+ ]
2750
+ ],
2751
+ "right_ranges": [
2752
+ [
2753
+ 26,
2754
+ 31
2755
+ ],
2756
+ [
2757
+ 34,
2758
+ 44
2759
+ ],
2760
+ [
2761
+ 136,
2762
+ 141
2763
+ ],
2764
+ [
2765
+ 180,
2766
+ 185
2767
+ ],
2768
+ [
2769
+ 281,
2770
+ 293
2771
+ ],
2772
+ [
2773
+ 325,
2774
+ 339
2775
+ ],
2776
+ [
2777
+ 346,
2778
+ 352
2779
+ ],
2780
+ [
2781
+ 358,
2782
+ 367
2783
+ ],
2784
+ [
2785
+ 397,
2786
+ 404
2787
+ ],
2788
+ [
2789
+ 409,
2790
+ 417
2791
+ ],
2792
+ [
2793
+ 426,
2794
+ 434
2795
+ ],
2796
+ [
2797
+ 463,
2798
+ 483
2799
+ ],
2800
+ [
2801
+ 498,
2802
+ 504
2803
+ ],
2804
+ [
2805
+ 510,
2806
+ 519
2807
+ ],
2808
+ [
2809
+ 532,
2810
+ 550
2811
+ ],
2812
+ [
2813
+ 553,
2814
+ 564
2815
+ ],
2816
+ [
2817
+ 586,
2818
+ 592
2819
+ ],
2820
+ [
2821
+ 598,
2822
+ 605
2823
+ ],
2824
+ [
2825
+ 618,
2826
+ 634
2827
+ ],
2828
+ [
2829
+ 676,
2830
+ 688
2831
+ ],
2832
+ [
2833
+ 699,
2834
+ 705
2835
+ ],
2836
+ [
2837
+ 711,
2838
+ 718
2839
+ ],
2840
+ [
2841
+ 731,
2842
+ 748
2843
+ ],
2844
+ [
2845
+ 752,
2846
+ 762
2847
+ ],
2848
+ [
2849
+ 771,
2850
+ 777
2851
+ ],
2852
+ [
2853
+ 782,
2854
+ 789
2855
+ ],
2856
+ [
2857
+ 798,
2858
+ 813
2859
+ ],
2860
+ [
2861
+ 831,
2862
+ 836
2863
+ ]
2864
+ ]
2865
+ },
2866
+ {
2867
+ "filename": "tests/models/gpt_neo/test_modeling_gpt_neo.py",
2868
+ "left_ranges": [
2869
+ [
2870
+ 458,
2871
+ 464
2872
+ ]
2873
+ ],
2874
+ "right_ranges": [
2875
+ [
2876
+ 458,
2877
+ 464
2878
+ ]
2879
+ ]
2880
+ }
2881
+ ]
2882
+ }
2883
+ ]
2884
+ }
2885
+ ],
2886
+ "duplicate_issues": [],
2887
+ "duplicate_prs": [
2888
+ {
2889
+ "cluster_id": "cluster-41115-3",
2890
+ "canonical_pr_number": 41116,
2891
+ "duplicate_pr_numbers": [
2892
+ 45613
2893
+ ],
2894
+ "target_issue_number": 41115,
2895
+ "reason": "PRs in cluster-41115-3 are treated as duplicates because they converge on issue #41115 with closing_reference evidence."
2896
+ },
2897
+ {
2898
+ "cluster_id": "cluster-41211-3",
2899
+ "canonical_pr_number": 44339,
2900
+ "duplicate_pr_numbers": [
2901
+ 41356
2902
+ ],
2903
+ "target_issue_number": 41211,
2904
+ "reason": "PRs in cluster-41211-3 are treated as duplicates because they converge on issue #41211 with closing_reference evidence."
2905
+ },
2906
+ {
2907
+ "cluster_id": "cluster-43240-3",
2908
+ "canonical_pr_number": 43251,
2909
+ "duplicate_pr_numbers": [
2910
+ 43254
2911
+ ],
2912
+ "target_issue_number": 43240,
2913
+ "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence."
2914
+ },
2915
+ {
2916
+ "cluster_id": "cluster-43698-3",
2917
+ "canonical_pr_number": 43779,
2918
+ "duplicate_pr_numbers": [
2919
+ 43816
2920
+ ],
2921
+ "target_issue_number": 43698,
2922
+ "reason": "PRs in cluster-43698-3 are treated as duplicates because they converge on issue #43698 with closing_reference, shared_issue_target evidence."
2923
+ },
2924
+ {
2925
+ "cluster_id": "cluster-43979-11",
2926
+ "canonical_pr_number": 44007,
2927
+ "duplicate_pr_numbers": [
2928
+ 43996,
2929
+ 44013,
2930
+ 44044,
2931
+ 44066,
2932
+ 44072,
2933
+ 44085,
2934
+ 44129,
2935
+ 44154,
2936
+ 44722
2937
+ ],
2938
+ "target_issue_number": 43979,
2939
+ "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence."
2940
+ },
2941
+ {
2942
+ "cluster_id": "cluster-44018-2",
2943
+ "canonical_pr_number": 44068,
2944
+ "duplicate_pr_numbers": [
2945
+ 44018
2946
+ ],
2947
+ "target_issue_number": null,
2948
+ "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence."
2949
+ },
2950
+ {
2951
+ "cluster_id": "cluster-45081-3",
2952
+ "canonical_pr_number": 45317,
2953
+ "duplicate_pr_numbers": [
2954
+ 45086
2955
+ ],
2956
+ "target_issue_number": 45081,
2957
+ "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence."
2958
+ },
2959
+ {
2960
+ "cluster_id": "cluster-45561-3",
2961
+ "canonical_pr_number": 45645,
2962
+ "duplicate_pr_numbers": [
2963
+ 45639
2964
+ ],
2965
+ "target_issue_number": 45561,
2966
+ "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference, shared_issue_target, soft_similarity evidence."
2967
+ },
2968
+ {
2969
+ "cluster_id": "cluster-45647-3",
2970
+ "canonical_pr_number": 45738,
2971
+ "duplicate_pr_numbers": [
2972
+ 45737
2973
+ ],
2974
+ "target_issue_number": 45647,
2975
+ "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence."
2976
+ },
2977
+ {
2978
+ "cluster_id": "cluster-45987-3",
2979
+ "canonical_pr_number": 46173,
2980
+ "duplicate_pr_numbers": [
2981
+ 46474
2982
+ ],
2983
+ "target_issue_number": 45987,
2984
+ "reason": "PRs in cluster-45987-3 are treated as duplicates because they converge on issue #45987 with closing_reference, shared_issue_target evidence."
2985
+ },
2986
+ {
2987
+ "cluster_id": "cluster-46159-3",
2988
+ "canonical_pr_number": 46174,
2989
+ "duplicate_pr_numbers": [
2990
+ 46780
2991
+ ],
2992
+ "target_issue_number": 46159,
2993
+ "reason": "PRs in cluster-46159-3 are treated as duplicates because they converge on issue #46159 with closing_reference evidence."
2994
+ },
2995
+ {
2996
+ "cluster_id": "cluster-46305-3",
2997
+ "canonical_pr_number": 46546,
2998
+ "duplicate_pr_numbers": [
2999
+ 46310
3000
+ ],
3001
+ "target_issue_number": 46305,
3002
+ "reason": "PRs in cluster-46305-3 are treated as duplicates because they converge on issue #46305 with closing_reference, shared_issue_target, soft_similarity evidence."
3003
+ },
3004
+ {
3005
+ "cluster_id": "cluster-46682-3",
3006
+ "canonical_pr_number": 46788,
3007
+ "duplicate_pr_numbers": [
3008
+ 46757
3009
+ ],
3010
+ "target_issue_number": 46682,
3011
+ "reason": "PRs in cluster-46682-3 are treated as duplicates because they converge on issue #46682 with closing_reference, shared_issue_target evidence."
3012
+ },
3013
+ {
3014
+ "cluster_id": "cluster-46752-7",
3015
+ "canonical_pr_number": 46793,
3016
+ "duplicate_pr_numbers": [
3017
+ 46753,
3018
+ 46778,
3019
+ 46789,
3020
+ 46791,
3021
+ 46794
3022
+ ],
3023
+ "target_issue_number": 46752,
3024
+ "reason": "PRs in cluster-46752-7 are treated as duplicates because they converge on issue #46752 with closing_reference, shared_issue_target, soft_similarity evidence."
3025
+ }
3026
+ ],
3027
+ "best_issue": {
3028
+ "cluster_id": "cluster-46752-7",
3029
+ "issue_number": 46752,
3030
+ "reason": "Issue #46752 is the strongest global issue candidate because it is open, belongs to a cluster with 7 artifacts, and carries 0 discussion comments plus 16 inbound references.",
3031
+ "score": 186.0
3032
+ },
3033
+ "best_pr": {
3034
+ "cluster_id": "cluster-43979-11",
3035
+ "pr_number": 44007,
3036
+ "reason": "PR #44007 is the strongest global PR candidate because it is open, belongs to a cluster with 11 artifacts, links to 1 issues, and carries 2 review/discussion events.",
3037
+ "score": 238.886
3038
+ }
3039
+ }
snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.llm-reviews.json ADDED
The diff for this file is too large to render. See raw diff
 
snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/manifest.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "analysis_id": "hybrid-model-20260622t000017z",
3
+ "artifacts": {
4
+ "hybrid": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.json",
5
+ "hybrid_reviews": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.llm-reviews.json"
6
+ },
7
+ "channel": "canonical",
8
+ "model": null,
9
+ "published_at": "2026-06-22T00:04:30Z",
10
+ "repo": "huggingface/transformers",
11
+ "schema_version": 1,
12
+ "snapshot_id": "20260622T000017Z",
13
+ "variant": "hybrid"
14
+ }
snapshots/20260622T000017Z/manifest.json CHANGED
@@ -32,6 +32,24 @@
32
  "timeline_events": 240
33
  },
34
  "extracted_at": "2026-06-22T00:00:17Z",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
35
  "repo": "huggingface/transformers",
36
  "snapshot_id": "20260622T000017Z",
37
  "watermark": {
 
32
  "timeline_events": 240
33
  },
34
  "extracted_at": "2026-06-22T00:00:17Z",
35
+ "published_analysis": {
36
+ "canonical_analysis_id": "hybrid-model-20260622t000017z",
37
+ "runs": {
38
+ "hybrid-model-20260622t000017z": {
39
+ "analysis_id": "hybrid-model-20260622t000017z",
40
+ "artifacts": {
41
+ "hybrid": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.json",
42
+ "hybrid_reviews": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/analysis-report-hybrid.llm-reviews.json"
43
+ },
44
+ "channel": "canonical",
45
+ "manifest_path": "snapshots/20260622T000017Z/analysis-runs/hybrid-model-20260622t000017z/manifest.json",
46
+ "model": null,
47
+ "published_at": "2026-06-22T00:04:30Z",
48
+ "variant": "hybrid"
49
+ }
50
+ },
51
+ "schema_version": 1
52
+ },
53
  "repo": "huggingface/transformers",
54
  "snapshot_id": "20260622T000017Z",
55
  "watermark": {