Initial model

1_Dense/config.json

@@ -0,0 +1 @@
+{"in_features": 50, "out_features": 128, "bias": false, "activation_function": "torch.nn.modules.linear.Identity"}

1_Dense/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bce499da1d2b9a1bfdea8171778d4e56cb4afd6fe8fdc72a4937651eb86dc77
+size 25688

README.md

@@ -0,0 +1,1090 @@
+---
+language:
+- en
+tags:
+- ColBERT
+- PyLate
+- sentence-transformers
+- sentence-similarity
+- feature-extraction
+- generated_from_trainer
+- dataset_size:640000
+- loss:Distillation
+datasets:
+- lightonai/ms-marco-en-bge-gemma
+pipeline_tag: sentence-similarity
+library_name: PyLate
+metrics:
+- MaxSim_accuracy@1
+- MaxSim_accuracy@3
+- MaxSim_accuracy@5
+- MaxSim_accuracy@10
+- MaxSim_precision@1
+- MaxSim_precision@3
+- MaxSim_precision@5
+- MaxSim_precision@10
+- MaxSim_recall@1
+- MaxSim_recall@3
+- MaxSim_recall@5
+- MaxSim_recall@10
+- MaxSim_ndcg@10
+- MaxSim_mrr@10
+- MaxSim_map@100
+model-index:
+- name: ColBERT MUVERA Femto
+  results:
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoClimateFEVER
+      type: NanoClimateFEVER
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.14
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.32
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.36
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.52
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.14
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.11333333333333333
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.07600000000000001
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.05600000000000001
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.085
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.165
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.19166666666666668
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.25233333333333335
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.19115874409066272
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.2408333333333333
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.1462389973257929
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoDBPedia
+      type: NanoDBPedia
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.7
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.82
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.82
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.84
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.7
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.5933333333333333
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.548
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.456
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.0728506527388449
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.13076941366456654
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.17827350013263704
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.2781635119304686
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.5510945084552747
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.7555555555555555
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.39128533545834626
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoFEVER
+      type: NanoFEVER
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.62
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.76
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.84
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.86
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.62
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.26666666666666666
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.184
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.09399999999999999
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.5766666666666667
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.7366666666666667
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.83
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.85
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.7249306483092258
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.6976666666666665
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.679664802101873
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoFiQA2018
+      type: NanoFiQA2018
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.28
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.34
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.44
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.48
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.28
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.13333333333333333
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.10800000000000001
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.062
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.13555555555555557
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.19755555555555557
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.2666349206349206
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.2994920634920635
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.2502784944505909
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.33252380952380944
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.20907273372726215
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoHotpotQA
+      type: NanoHotpotQA
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.76
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.84
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.9
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.92
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.76
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.36666666666666664
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.252
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.136
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.38
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.55
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.63
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.68
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.6514325561331983
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.8098333333333333
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.5738665952275315
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoMSMARCO
+      type: NanoMSMARCO
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.32
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.48
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.6
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.7
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.32
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.16
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.12000000000000002
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.07
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.32
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.48
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.6
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.7
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.4946222844793249
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.43052380952380953
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.4408050908765128
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoNFCorpus
+      type: NanoNFCorpus
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.32
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.42
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.52
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.62
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.32
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.2866666666666666
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.26799999999999996
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.21000000000000005
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.01921769353070746
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.03782391241260524
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.05411010345369293
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.09349869834347448
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.2481257474345093
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.3995793650793651
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.08737709081330662
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoNQ
+      type: NanoNQ
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.28
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.52
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.58
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.74
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.28
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.1733333333333333
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.12000000000000002
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.07600000000000001
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.26
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.49
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.56
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.7
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.4828411530427104
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.4289603174603174
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.41150699780701017
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoQuoraRetrieval
+      type: NanoQuoraRetrieval
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.74
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.84
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.88
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.9
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.74
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.30666666666666664
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.21199999999999997
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.11599999999999998
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.674
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.784
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.8413333333333333
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.8626666666666667
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.8016479127266055
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.7995238095238095
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.7733654571274
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoSCIDOCS
+      type: NanoSCIDOCS
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.3
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.44
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.52
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.62
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.3
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.18
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.14400000000000002
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.092
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.061000000000000006
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.11100000000000002
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.14700000000000002
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.18799999999999997
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.198564235862039
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.3978253968253968
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.13670583023266375
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoArguAna
+      type: NanoArguAna
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.14
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.24
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.28
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.36
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.14
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.07999999999999999
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.05600000000000001
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.036000000000000004
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.14
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.24
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.28
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.36
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.2444065884095295
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.20804761904761904
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.21989999402599436
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoSciFact
+      type: NanoSciFact
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.36
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.5
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.6
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.62
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.36
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.18666666666666668
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.14
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.07400000000000001
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.325
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.49
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.59
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.62
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.4856083424090788
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.44449999999999995
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.44726079800650204
+      name: Maxsim Map@100
+  - task:
+      type: py-late-information-retrieval
+      name: Py Late Information Retrieval
+    dataset:
+      name: NanoTouche2020
+      type: NanoTouche2020
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.6530612244897959
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.9591836734693877
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.9795918367346939
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 1.0
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.6530612244897959
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.6054421768707483
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.5673469387755103
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.45918367346938777
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.04308959031413618
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.11831839494199368
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.1804772716223025
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.2842813442856462
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.5191595399345652
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.8064625850340136
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.32574548665687825
+      name: Maxsim Map@100
+  - task:
+      type: nano-beir
+      name: Nano BEIR
+    dataset:
+      name: NanoBEIR mean
+      type: NanoBEIR_mean
+    metrics:
+    - type: MaxSim_accuracy@1
+      value: 0.4317739403453689
+      name: Maxsim Accuracy@1
+    - type: MaxSim_accuracy@3
+      value: 0.5753218210361067
+      name: Maxsim Accuracy@3
+    - type: MaxSim_accuracy@5
+      value: 0.6399686028257457
+      name: Maxsim Accuracy@5
+    - type: MaxSim_accuracy@10
+      value: 0.7061538461538461
+      name: Maxsim Accuracy@10
+    - type: MaxSim_precision@1
+      value: 0.4317739403453689
+      name: Maxsim Precision@1
+    - type: MaxSim_precision@3
+      value: 0.26554683411826263
+      name: Maxsim Precision@3
+    - type: MaxSim_precision@5
+      value: 0.2150266875981162
+      name: Maxsim Precision@5
+    - type: MaxSim_precision@10
+      value: 0.14901412872841444
+      name: Maxsim Precision@10
+    - type: MaxSim_recall@1
+      value: 0.2378753968312239
+      name: Maxsim Recall@1
+    - type: MaxSim_recall@3
+      value: 0.34854876486472214
+      name: Maxsim Recall@3
+    - type: MaxSim_recall@5
+      value: 0.41149967660335024
+      name: Maxsim Recall@5
+    - type: MaxSim_recall@10
+      value: 0.4744950475424349
+      name: Maxsim Recall@10
+    - type: MaxSim_ndcg@10
+      value: 0.44952851967210117
+      name: Maxsim Ndcg@10
+    - type: MaxSim_mrr@10
+      value: 0.5193719693005406
+      name: Maxsim Mrr@10
+    - type: MaxSim_map@100
+      value: 0.3725227084143902
+      name: Maxsim Map@100
+---
+
+# ColBERT MUVERA Femto
+
+This is a [PyLate](https://github.com/lightonai/pylate) model finetuned from [neuml/bert-hash-femto](https://huggingface.co/neuml/bert-hash-femto) on the [msmarco-en-bge-gemma unnormalized split](https://huggingface.co/datasets/lightonai/ms-marco-en-bge-gemma) dataset. It maps sentences & paragraphs to sequences of 50-dimensional dense vectors and can be used for semantic textual similarity using the MaxSim operator.
+
+This model is trained with un-normalized scores, making it compatible with [MUVERA fixed-dimensional encoding](https://arxiv.org/abs/2405.19504).
+
+## Usage (txtai)
+
+This model can be used to build embeddings databases with [txtai](https://github.com/neuml/txtai) for semantic search and/or as a knowledge source for retrieval augmented generation (RAG).
+
+_Note: txtai 9.0+ is required for late interaction model support_
+
+```python
+import txtai
+
+embeddings = txtai.Embeddings(
+  sparse="neuml/colbert-muvera-femto",
+  content=True
+)
+embeddings.index(documents())
+
+# Run a query
+embeddings.search("query to run")
+```
+
+Late interaction models excel as reranker pipelines.
+
+```python
+from txtai.pipeline import Reranker, Similarity
+
+similarity = Similarity(path="neuml/colbert-muvera-femto", lateencode=True)
+ranker = Reranker(embeddings, similarity)
+ranker("query to run")
+```
+
+## Usage (PyLate)
+
+Alternatively, the model can be loaded with [PyLate](https://github.com/lightonai/pylate).
+
+```python
+from pylate import rank, models
+
+queries = [
+    "query A",
+    "query B",
+]
+
+documents = [
+    ["document A", "document B"],
+    ["document 1", "document C", "document B"],
+]
+
+documents_ids = [
+    [1, 2],
+    [1, 3, 2],
+]
+
+model = models.ColBERT(
+    model_name_or_path="neuml/colbert-muvera-femto",
+)
+
+queries_embeddings = model.encode(
+    queries,
+    is_query=True,
+)
+
+documents_embeddings = model.encode(
+    documents,
+    is_query=False,
+)
+
+reranked_documents = rank.rerank(
+    documents_ids=documents_ids,
+    queries_embeddings=queries_embeddings,
+    documents_embeddings=documents_embeddings,
+)
+```
+
+### Full Model Architecture
+
+```
+ColBERT(
+  (0): Transformer({'max_seq_length': 299, 'do_lower_case': False}) with Transformer model: BertHashModel 
+  (1): Dense({'in_features': 50, 'out_features': 50, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
+)
+```
+
+## Evaluation
+
+### BEIR Subset
+
+The following table shows a subset of BEIR scored with the [txtai benchmarks script](https://github.com/neuml/txtai/blob/master/examples/benchmarks.py). 
+
+Scores reported are `ndcg@10` and grouped into the following three categories.
+
+#### FULL multi-vector maxsim
+
+| Model             | Parameters | NFCorpus | SciDocs  | SciFact | Average |
+|:------------------|:-----------|:---------|:---------|:--------|:--------|
+| [ColBERT v2](https://huggingface.co/colbert-ir/colbertv2.0) | 110M | 0.3165 | 0.1497 | 0.6456 | 0.3706 |
+| [**ColBERT MUVERA Femto**](https://huggingface.co/neuml/colbert-muvera-femto) | **0.2M** | **0.2513** | **0.0870** | **0.4710** | **0.2698** |
+| [ColBERT MUVERA Pico](https://huggingface.co/neuml/colbert-muvera-pico) | 0.4M | 0.3005 | 0.1117 | 0.6452 | 0.3525 |
+| [ColBERT MUVERA Nano](https://huggingface.co/neuml/colbert-muvera-nano) | 0.9M | 0.3180 | 0.1262 | 0.6576 | 0.3673 |
+| [ColBERT MUVERA Micro](https://huggingface.co/neuml/colbert-muvera-micro) | 4M | 0.3235 | 0.1244 | 0.6676 | 0.3718 |
+
+#### MUVERA encoding + maxsim re-ranking of the top 100 results per MUVERA paper
+
+| Model             | Parameters | NFCorpus | SciDocs  | SciFact | Average |
+|:------------------|:-----------|:---------|:---------|:--------|:--------|
+| [ColBERT v2](https://huggingface.co/colbert-ir/colbertv2.0) | 110M | 0.3025 | 0.1538 | 0.6278 | 0.3614 | 
+| [**ColBERT MUVERA Femto**](https://huggingface.co/neuml/colbert-muvera-femto) | **0.2M** | **0.2316** | **0.0858** | **0.4641** | **0.2605** |
+| [ColBERT MUVERA Pico](https://huggingface.co/neuml/colbert-muvera-pico) | 0.4M | 0.2821 | 0.1004 | 0.6090 | 0.3305 |
+| [ColBERT MUVERA Nano](https://huggingface.co/neuml/colbert-muvera-nano) | 0.9M | 0.2996 | 0.1201 | 0.6249 | 0.3482 |
+| [ColBERT MUVERA Micro](https://huggingface.co/neuml/colbert-muvera-micro) | 4M | 0.3095 | 0.1228 | 0.6464 | 0.3596 |
+
+#### MUVERA encoding only
+
+| Model             | Parameters | NFCorpus | SciDocs  | SciFact | Average |
+|:------------------|:-----------|:---------|:---------|:--------|:--------|
+| [ColBERT v2](https://huggingface.co/colbert-ir/colbertv2.0) | 110M | 0.2356 | 0.1229 | 0.5002 | 0.2862 |
+| [**ColBERT MUVERA Femto**](https://huggingface.co/neuml/colbert-muvera-femto) | **0.2M** | **0.1851** | **0.0411** | **0.3518** | **0.1927** |
+| [ColBERT MUVERA Pico](https://huggingface.co/neuml/colbert-muvera-pico) | 0.4M | 0.1926 | 0.0564 | 0.4424 | 0.2305 |
+| [ColBERT MUVERA Nano](https://huggingface.co/neuml/colbert-muvera-nano) | 0.9M | 0.2355 | 0.0807 | 0.4904 | 0.2689 |
+| [ColBERT MUVERA Micro](https://huggingface.co/neuml/colbert-muvera-micro) | 4M | 0.2348 | 0.0882 | 0.4875 | 0.2702 |
+
+_Note: The scores reported don't match scores reported in the respective papers due to different default settings in the txtai benchmark scripts._
+
+As noted earlier, models trained with min-max score normalization don't perform well with MUVERA encoding. See this [GitHub Issue](https://github.com/lightonai/pylate/issues/142) for more.
+
+**This model is only 250K parameters with a file size of 950K. Keeping that in mind, it's surprising how decent the scores are!**
+
+### Nano BEIR
+* Dataset: `NanoBEIR_mean`
+* Evaluated with <code>pylate.evaluation.nano_beir_evaluator.NanoBEIREvaluator</code>
+
+| Metric              | Value      |
+|:--------------------|:-----------|
+| MaxSim_accuracy@1   | 0.4318     |
+| MaxSim_accuracy@3   | 0.5753     |
+| MaxSim_accuracy@5   | 0.64       |
+| MaxSim_accuracy@10  | 0.7062     |
+| MaxSim_precision@1  | 0.4318     |
+| MaxSim_precision@3  | 0.2655     |
+| MaxSim_precision@5  | 0.215      |
+| MaxSim_precision@10 | 0.149      |
+| MaxSim_recall@1     | 0.2379     |
+| MaxSim_recall@3     | 0.3485     |
+| MaxSim_recall@5     | 0.4115     |
+| MaxSim_recall@10    | 0.4745     |
+| **MaxSim_ndcg@10**  | **0.4495** |
+| MaxSim_mrr@10       | 0.5194     |
+| MaxSim_map@100      | 0.3725     |
+
+## Training Details
+
+### Training Hyperparameters
+
+#### Non-Default Hyperparameters
+
+- `eval_strategy`: steps
+- `per_device_train_batch_size`: 32
+- `learning_rate`: 0.0003
+- `num_train_epochs`: 1
+- `warmup_ratio`: 0.05
+- `fp16`: True
+
+#### All Hyperparameters
+<details><summary>Click to expand</summary>
+
+- `overwrite_output_dir`: False
+- `do_predict`: False
+- `eval_strategy`: steps
+- `prediction_loss_only`: True
+- `per_device_train_batch_size`: 32
+- `per_device_eval_batch_size`: 8
+- `per_gpu_train_batch_size`: None
+- `per_gpu_eval_batch_size`: None
+- `gradient_accumulation_steps`: 1
+- `eval_accumulation_steps`: None
+- `torch_empty_cache_steps`: None
+- `learning_rate`: 0.0003
+- `weight_decay`: 0.0
+- `adam_beta1`: 0.9
+- `adam_beta2`: 0.999
+- `adam_epsilon`: 1e-08
+- `max_grad_norm`: 1.0
+- `num_train_epochs`: 1
+- `max_steps`: -1
+- `lr_scheduler_type`: linear
+- `lr_scheduler_kwargs`: {}
+- `warmup_ratio`: 0.05
+- `warmup_steps`: 0
+- `log_level`: passive
+- `log_level_replica`: warning
+- `log_on_each_node`: True
+- `logging_nan_inf_filter`: True
+- `save_safetensors`: True
+- `save_on_each_node`: False
+- `save_only_model`: False
+- `restore_callback_states_from_checkpoint`: False
+- `no_cuda`: False
+- `use_cpu`: False
+- `use_mps_device`: False
+- `seed`: 42
+- `data_seed`: None
+- `jit_mode_eval`: False
+- `bf16`: False
+- `fp16`: True
+- `fp16_opt_level`: O1
+- `half_precision_backend`: auto
+- `bf16_full_eval`: False
+- `fp16_full_eval`: False
+- `tf32`: None
+- `local_rank`: 0
+- `ddp_backend`: None
+- `tpu_num_cores`: None
+- `tpu_metrics_debug`: False
+- `debug`: []
+- `dataloader_drop_last`: False
+- `dataloader_num_workers`: 0
+- `dataloader_prefetch_factor`: None
+- `past_index`: -1
+- `disable_tqdm`: False
+- `remove_unused_columns`: True
+- `label_names`: None
+- `load_best_model_at_end`: False
+- `ignore_data_skip`: False
+- `fsdp`: []
+- `fsdp_min_num_params`: 0
+- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
+- `fsdp_transformer_layer_cls_to_wrap`: None
+- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
+- `parallelism_config`: None
+- `deepspeed`: None
+- `label_smoothing_factor`: 0.0
+- `optim`: adamw_torch_fused
+- `optim_args`: None
+- `adafactor`: False
+- `group_by_length`: False
+- `length_column_name`: length
+- `project`: huggingface
+- `trackio_space_id`: trackio
+- `ddp_find_unused_parameters`: None
+- `ddp_bucket_cap_mb`: None
+- `ddp_broadcast_buffers`: False
+- `dataloader_pin_memory`: True
+- `dataloader_persistent_workers`: False
+- `skip_memory_metrics`: True
+- `use_legacy_prediction_loop`: False
+- `push_to_hub`: False
+- `resume_from_checkpoint`: None
+- `hub_model_id`: None
+- `hub_strategy`: every_save
+- `hub_private_repo`: None
+- `hub_always_push`: False
+- `hub_revision`: None
+- `gradient_checkpointing`: False
+- `gradient_checkpointing_kwargs`: None
+- `include_inputs_for_metrics`: False
+- `include_for_metrics`: []
+- `eval_do_concat_batches`: True
+- `fp16_backend`: auto
+- `push_to_hub_model_id`: None
+- `push_to_hub_organization`: None
+- `mp_parameters`: 
+- `auto_find_batch_size`: False
+- `full_determinism`: False
+- `torchdynamo`: None
+- `ray_scope`: last
+- `ddp_timeout`: 1800
+- `torch_compile`: False
+- `torch_compile_backend`: None
+- `torch_compile_mode`: None
+- `include_tokens_per_second`: False
+- `include_num_input_tokens_seen`: no
+- `neftune_noise_alpha`: None
+- `optim_target_modules`: None
+- `batch_eval_metrics`: False
+- `eval_on_start`: False
+- `use_liger_kernel`: False
+- `liger_kernel_config`: None
+- `eval_use_gather_object`: False
+- `average_tokens_across_devices`: True
+- `prompts`: None
+- `batch_sampler`: batch_sampler
+- `multi_dataset_batch_sampler`: proportional
+
+</details>
+
+### Framework Versions
+- Python: 3.10.18
+- Sentence Transformers: 4.0.2
+- PyLate: 1.3.2
+- Transformers: 4.57.0
+- PyTorch: 2.8.0+cu128
+- Accelerate: 1.10.1
+- Datasets: 4.1.1
+- Tokenizers: 0.22.1
+
+## Citation
+
+### BibTeX
+
+#### Sentence Transformers
+```bibtex
+@inproceedings{reimers-2019-sentence-bert,
+    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
+    author = "Reimers, Nils and Gurevych, Iryna",
+    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
+    month = "11",
+    year = "2019",
+    publisher = "Association for Computational Linguistics",
+    url = "https://arxiv.org/abs/1908.10084"
+}
+```
+
+#### PyLate
+```bibtex
+@misc{PyLate,
+title={PyLate: Flexible Training and Retrieval for Late Interaction Models},
+author={Chaffin, Antoine and Sourty, Raphaël},
+url={https://github.com/lightonai/pylate},
+year={2024}
+}
+```

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "[D] ": 30523,
+  "[Q] ": 30522
+}

config.json

@@ -0,0 +1,31 @@
+{
+  "architectures": [
+    "BertHashModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "auto_map": {
+    "AutoConfig": "configuration_bert_hash.BertHashConfig",
+    "AutoModel": "modeling_bert_hash.BertHashModel",
+    "AutoModelForMaskedLM": "modeling_bert_hash.BertHashForMaskedLM",
+    "AutoModelForSequenceClassification": "modeling_bert_hash.BertHashForSequenceClassification"
+  },
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 50,
+  "initializer_range": 0.02,
+  "intermediate_size": 200,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert_hash",
+  "num_attention_heads": 2,
+  "num_hidden_layers": 2,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "projections": 5,
+  "transformers_version": "4.57.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30524
+}

config_sentence_transformers.json

@@ -0,0 +1,50 @@
+{
+  "__version__": {
+    "sentence_transformers": "4.0.2",
+    "transformers": "4.57.0",
+    "pytorch": "2.8.0+cu128"
+  },
+  "prompts": {},
+  "default_prompt_name": null,
+  "similarity_fn_name": "MaxSim",
+  "query_prefix": "[Q] ",
+  "document_prefix": "[D] ",
+  "query_length": 32,
+  "document_length": 300,
+  "attend_to_expansion_tokens": false,
+  "skiplist_words": [
+    "!",
+    "\"",
+    "#",
+    "$",
+    "%",
+    "&",
+    "'",
+    "(",
+    ")",
+    "*",
+    "+",
+    ",",
+    "-",
+    ".",
+    "/",
+    ":",
+    ";",
+    "<",
+    "=",
+    ">",
+    "?",
+    "@",
+    "[",
+    "\\",
+    "]",
+    "^",
+    "_",
+    "`",
+    "{",
+    "|",
+    "}",
+    "~"
+  ],
+  "do_query_expansion": true
+}

configuration_bert_hash.py

@@ -0,0 +1,14 @@
+from transformers.models.bert.configuration_bert import BertConfig
+
+
+class BertHashConfig(BertConfig):
+    """
+    Extension of Bert configuration to add projections parameter.
+    """
+
+    model_type = "bert_hash"
+
+    def __init__(self, projections=5, **kwargs):
+        super().__init__(**kwargs)
+
+        self.projections = projections

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c394ea7e54b710573bc1d8a8ec83a5689db3d0d7b97c47718da535332eafa642
+size 974552

modeling_bert_hash.py

@@ -0,0 +1,519 @@
+from typing import Optional, Union
+
+import torch
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+
+from transformers.cache_utils import Cache
+from transformers.models.bert.modeling_bert import BertEncoder, BertPooler, BertPreTrainedModel, BertOnlyMLMHead
+from transformers.modeling_attn_mask_utils import _prepare_4d_attention_mask_for_sdpa, _prepare_4d_causal_attention_mask_for_sdpa
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPoolingAndCrossAttentions,
+    MaskedLMOutput,
+    SequenceClassifierOutput,
+)
+from transformers.utils import auto_docstring, logging
+
+from .configuration_bert_hash import BertHashConfig
+
+logger = logging.get_logger(__name__)
+
+
+class BertHashTokens(nn.Module):
+    """
+    Module that embeds token vocabulary to an intermediate embeddings layer then projects those embeddings to the
+    hidden size.
+
+    The number of projections is like a hash. Setting the projections parameter to 5 is like generating a
+    160-bit hash (5 x float32) for each token. That hash is then projected to the hidden size.
+
+    This significantly reduces the number of parameters necessary for token embeddings.
+
+    For example:
+        Standard token embeddings:
+          30,522 (vocab size) x 768 (hidden size) = 23,440,896 parameters
+          23,440,896 x 4 (float32) = 93,763,584 bytes
+
+        Hash token embeddings:
+          30,522 (vocab size) x 5 (hash buckets) + 5 x 768 (projection matrix)= 156,450 parameters
+          156,450 x 4 (float32) = 625,800 bytes    
+    """
+
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+
+        # Token embeddings
+        self.embeddings = nn.Embedding(config.vocab_size, config.projections, padding_idx=config.pad_token_id)
+
+        # Token embeddings projections
+        self.projections = nn.Linear(config.projections, config.hidden_size)
+
+    def forward(self, input_ids):
+        # Project embeddings to hidden size
+        return self.projections(self.embeddings(input_ids))
+
+
+class BertHashEmbeddings(nn.Module):
+    """Construct the embeddings from word, position and token_type embeddings."""
+
+    def __init__(self, config):
+        super().__init__()
+        self.word_embeddings = BertHashTokens(config)
+        self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.hidden_size)
+        self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)
+
+        # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
+        # any TensorFlow checkpoint file
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
+        self.position_embedding_type = getattr(config, "position_embedding_type", "absolute")
+        self.register_buffer(
+            "position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)), persistent=False
+        )
+        self.register_buffer(
+            "token_type_ids", torch.zeros(self.position_ids.size(), dtype=torch.long), persistent=False
+        )
+
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        token_type_ids: Optional[torch.LongTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        past_key_values_length: int = 0,
+    ) -> torch.Tensor:
+        if input_ids is not None:
+            input_shape = input_ids.size()
+        else:
+            input_shape = inputs_embeds.size()[:-1]
+
+        seq_length = input_shape[1]
+
+        if position_ids is None:
+            position_ids = self.position_ids[:, past_key_values_length : seq_length + past_key_values_length]
+
+        # Setting the token_type_ids to the registered buffer in constructor where it is all zeros, which usually occurs
+        # when its auto-generated, registered buffer helps users when tracing the model without passing token_type_ids, solves
+        # issue #5664
+        if token_type_ids is None:
+            if hasattr(self, "token_type_ids"):
+                buffered_token_type_ids = self.token_type_ids[:, :seq_length]
+                buffered_token_type_ids_expanded = buffered_token_type_ids.expand(input_shape[0], seq_length)
+                token_type_ids = buffered_token_type_ids_expanded
+            else:
+                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=self.position_ids.device)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.word_embeddings(input_ids)
+        token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+        embeddings = inputs_embeds + token_type_embeddings
+        if self.position_embedding_type == "absolute":
+            position_embeddings = self.position_embeddings(position_ids)
+            embeddings += position_embeddings
+        embeddings = self.LayerNorm(embeddings)
+        embeddings = self.dropout(embeddings)
+        return embeddings
+
+
+@auto_docstring(
+    custom_intro="""
+    The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of
+    cross-attention is added between the self-attention layers, following the architecture described in [Attention is
+    all you need](https://huggingface.co/papers/1706.03762) by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit,
+    Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.
+
+    To behave as an decoder the model needs to be initialized with the `is_decoder` argument of the configuration set
+    to `True`. To be used in a Seq2Seq model, the model needs to initialized with both `is_decoder` argument and
+    `add_cross_attention` set to `True`; an `encoder_hidden_states` is then expected as an input to the forward pass.
+    """
+)
+class BertHashModel(BertPreTrainedModel):
+    config_class = BertHashConfig
+
+    _no_split_modules = ["BertEmbeddings", "BertLayer"]
+
+    def __init__(self, config, add_pooling_layer=True):
+        r"""
+        add_pooling_layer (bool, *optional*, defaults to `True`):
+            Whether to add a pooling layer
+        """
+        super().__init__(config)
+        self.config = config
+
+        self.embeddings = BertHashEmbeddings(config)
+        self.encoder = BertEncoder(config)
+
+        self.pooler = BertPooler(config) if add_pooling_layer else None
+
+        self.attn_implementation = config._attn_implementation
+        self.position_embedding_type = config.position_embedding_type
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embeddings.word_embeddings.embeddings
+
+    def set_input_embeddings(self, value):
+        self.embeddings.word_embeddings.embeddings = value
+
+    def _prune_heads(self, heads_to_prune):
+        """
+        Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
+        class PreTrainedModel
+        """
+        for layer, heads in heads_to_prune.items():
+            self.encoder.layer[layer].attention.prune_heads(heads)
+
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        encoder_hidden_states: Optional[torch.Tensor] = None,
+        encoder_attention_mask: Optional[torch.Tensor] = None,
+        past_key_values: Optional[list[torch.FloatTensor]] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        cache_position: Optional[torch.Tensor] = None,
+    ) -> Union[tuple[torch.Tensor], BaseModelOutputWithPoolingAndCrossAttentions]:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if self.config.is_decoder:
+            use_cache = use_cache if use_cache is not None else self.config.use_cache
+        else:
+            use_cache = False
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
+            input_shape = input_ids.size()
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        batch_size, seq_length = input_shape
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        past_key_values_length = 0
+        if past_key_values is not None:
+            past_key_values_length = (
+                past_key_values[0][0].shape[-2]
+                if not isinstance(past_key_values, Cache)
+                else past_key_values.get_seq_length()
+            )
+
+        if token_type_ids is None:
+            if hasattr(self.embeddings, "token_type_ids"):
+                buffered_token_type_ids = self.embeddings.token_type_ids[:, :seq_length]
+                buffered_token_type_ids_expanded = buffered_token_type_ids.expand(batch_size, seq_length)
+                token_type_ids = buffered_token_type_ids_expanded
+            else:
+                token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
+
+        embedding_output = self.embeddings(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            token_type_ids=token_type_ids,
+            inputs_embeds=inputs_embeds,
+            past_key_values_length=past_key_values_length,
+        )
+
+        if attention_mask is None:
+            attention_mask = torch.ones((batch_size, seq_length + past_key_values_length), device=device)
+
+        use_sdpa_attention_masks = (
+            self.attn_implementation == "sdpa"
+            and self.position_embedding_type == "absolute"
+            and head_mask is None
+            and not output_attentions
+        )
+
+        # Expand the attention mask
+        if use_sdpa_attention_masks and attention_mask.dim() == 2:
+            # Expand the attention mask for SDPA.
+            # [bsz, seq_len] -> [bsz, 1, seq_len, seq_len]
+            if self.config.is_decoder:
+                extended_attention_mask = _prepare_4d_causal_attention_mask_for_sdpa(
+                    attention_mask,
+                    input_shape,
+                    embedding_output,
+                    past_key_values_length,
+                )
+            else:
+                extended_attention_mask = _prepare_4d_attention_mask_for_sdpa(
+                    attention_mask, embedding_output.dtype, tgt_len=seq_length
+                )
+        else:
+            # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+            # ourselves in which case we just need to make it broadcastable to all heads.
+            extended_attention_mask = self.get_extended_attention_mask(attention_mask, input_shape)
+
+        # If a 2D or 3D attention mask is provided for the cross-attention
+        # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
+        if self.config.is_decoder and encoder_hidden_states is not None:
+            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
+            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
+            if encoder_attention_mask is None:
+                encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
+
+            if use_sdpa_attention_masks and encoder_attention_mask.dim() == 2:
+                # Expand the attention mask for SDPA.
+                # [bsz, seq_len] -> [bsz, 1, seq_len, seq_len]
+                encoder_extended_attention_mask = _prepare_4d_attention_mask_for_sdpa(
+                    encoder_attention_mask, embedding_output.dtype, tgt_len=seq_length
+                )
+            else:
+                encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
+        else:
+            encoder_extended_attention_mask = None
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+        encoder_outputs = self.encoder(
+            embedding_output,
+            attention_mask=extended_attention_mask,
+            head_mask=head_mask,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_extended_attention_mask,
+            past_key_values=past_key_values,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            cache_position=cache_position,
+        )
+        sequence_output = encoder_outputs[0]
+        pooled_output = self.pooler(sequence_output) if self.pooler is not None else None
+
+        if not return_dict:
+            return (sequence_output, pooled_output) + encoder_outputs[1:]
+
+        return BaseModelOutputWithPoolingAndCrossAttentions(
+            last_hidden_state=sequence_output,
+            pooler_output=pooled_output,
+            past_key_values=encoder_outputs.past_key_values,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+            cross_attentions=encoder_outputs.cross_attentions,
+        )
+
+
+@auto_docstring
+class BertHashForMaskedLM(BertPreTrainedModel):
+    _tied_weights_keys = ["predictions.decoder.bias", "cls.predictions.decoder.weight"]
+    config_class = BertHashConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        if config.is_decoder:
+            logger.warning(
+                "If you want to use `BertForMaskedLM` make sure `config.is_decoder=False` for "
+                "bi-directional self-attention."
+            )
+
+        self.bert = BertHashModel(config, add_pooling_layer=False)
+        self.cls = BertOnlyMLMHead(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        encoder_hidden_states: Optional[torch.Tensor] = None,
+        encoder_attention_mask: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[tuple[torch.Tensor], MaskedLMOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
+            config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
+            loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
+        """
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.bert(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_attention_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+        prediction_scores = self.cls(sequence_output)
+
+        masked_lm_loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()  # -100 index = padding token
+            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
+
+        if not return_dict:
+            output = (prediction_scores,) + outputs[2:]
+            return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output
+
+        return MaskedLMOutput(
+            loss=masked_lm_loss,
+            logits=prediction_scores,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+    def prepare_inputs_for_generation(self, input_ids, attention_mask=None, **model_kwargs):
+        input_shape = input_ids.shape
+        effective_batch_size = input_shape[0]
+
+        #  add a dummy token
+        if self.config.pad_token_id is None:
+            raise ValueError("The PAD token should be defined for generation")
+
+        attention_mask = torch.cat([attention_mask, attention_mask.new_zeros((attention_mask.shape[0], 1))], dim=-1)
+        dummy_token = torch.full(
+            (effective_batch_size, 1), self.config.pad_token_id, dtype=torch.long, device=input_ids.device
+        )
+        input_ids = torch.cat([input_ids, dummy_token], dim=1)
+
+        return {"input_ids": input_ids, "attention_mask": attention_mask}
+
+    @classmethod
+    def can_generate(cls) -> bool:
+        """
+        Legacy correction: BertForMaskedLM can't call `generate()` from `GenerationMixin`, even though it has a
+        `prepare_inputs_for_generation` method.
+        """
+        return False
+
+
+@auto_docstring(
+    custom_intro="""
+    Bert Model transformer with a sequence classification/regression head on top (a linear layer on top of the pooled
+    output) e.g. for GLUE tasks.
+    """
+)
+class BertHashForSequenceClassification(BertPreTrainedModel):
+    config_class = BertHashConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.config = config
+
+        self.bert = BertHashModel(config)
+        classifier_dropout = (
+            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+        )
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        token_type_ids: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[tuple[torch.Tensor], SequenceClassifierOutput]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.bert(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        pooled_output = outputs[1]
+
+        pooled_output = self.dropout(pooled_output)
+        logits = self.classifier(pooled_output)
+
+        loss = None
+        if labels is not None:
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(logits, labels)
+        if not return_dict:
+            output = (logits,) + outputs[2:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )

modules.json

@@ -0,0 +1,14 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Dense",
+    "type": "pylate.models.Dense.Dense"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 299,
+  "do_lower_case": false
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "[MASK]",
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,72 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30522": {
+      "content": "[Q] ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30523": {
+      "content": "[D] ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[MASK]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}