Hugging Face's logo

redis
/
langcache-embed-v3

Sentence Similarity
sentence-transformers
Safetensors
English
modernbert
biencoder
text-classification
sentence-pair-classification
semantic-similarity
semantic-search
retrieval
reranking
Generated from Trainer
dataset_size:1451941
loss:MultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Model card Files
xet
Community
radoslavralev commited on
Commit
cf689fc
·
verified ·
1 Parent(s): 11e30c1

Add new SentenceTransformer model

Browse files
Files changed (2) hide show
  1. README.md +35 -372
  2. model.safetensors +1 -1
README.md CHANGED
@@ -12,8 +12,8 @@ tags:
12
  - retrieval
13
  - reranking
14
  - generated_from_trainer
15
- - dataset_size:483820
16
- - loss:MultipleNegativesSymmetricRankingLoss
17
  base_model: Alibaba-NLP/gte-modernbert-base
18
  widget:
19
  - source_sentence: 'See Precambrian time scale # Proposed Geologic timeline for another
@@ -85,22 +85,22 @@ model-index:
85
  type: train
86
  metrics:
87
  - type: cosine_accuracy@1
88
- value: 0.5978783286425633
89
  name: Cosine Accuracy@1
90
  - type: cosine_precision@1
91
- value: 0.5978783286425633
92
  name: Cosine Precision@1
93
  - type: cosine_recall@1
94
- value: 0.5765917883925028
95
  name: Cosine Recall@1
96
  - type: cosine_ndcg@10
97
- value: 0.7905393533594786
98
  name: Cosine Ndcg@10
99
  - type: cosine_mrr@1
100
- value: 0.5978783286425633
101
  name: Cosine Mrr@1
102
  - type: cosine_map@100
103
- value: 0.7375956597574003
104
  name: Cosine Map@100
105
  ---
106
 
@@ -165,9 +165,9 @@ print(embeddings.shape)
165
  # Get the similarity scores for the embeddings
166
  similarities = model.similarity(embeddings, embeddings)
167
  print(similarities)
168
- # tensor([[1.0000, 0.9922, 0.0547],
169
- # [0.9922, 1.0000, 0.0449],
170
- # [0.0547, 0.0449, 1.0000]], dtype=torch.bfloat16)
171
  ```
172
 
173
  <!--
@@ -205,12 +205,12 @@ You can finetune this model on your own dataset.
205
 
206
  | Metric | Value |
207
  |:-------------------|:-----------|
208
- | cosine_accuracy@1 | 0.5979 |
209
- | cosine_precision@1 | 0.5979 |
210
- | cosine_recall@1 | 0.5766 |
211
- | **cosine_ndcg@10** | **0.7905** |
212
- | cosine_mrr@1 | 0.5979 |
213
- | cosine_map@100 | 0.7376 |
214
 
215
  <!--
216
  ## Bias, Risks and Limitations
@@ -244,12 +244,11 @@ You can finetune this model on your own dataset.
244
  | <code>The newer Punts are still very much in existence today and race in the same fleets as the older boats .</code> | <code>The newer punts are still very much in existence today and run in the same fleets as the older boats .</code> | <code>1</code> |
245
  | <code>After losing his second election , he resigned as opposition leader and was replaced by Geoff Pearsall .</code> | <code>Max Bingham resigned as opposition leader after losing his second election , and was replaced by Geoff Pearsall .</code> | <code>1</code> |
246
  | <code>The 12F was officially homologated on August 21 , 1929 and exhibited at the Paris Salon in 1930 .</code> | <code>The 12F was officially homologated on 21 August 1929 and displayed at the 1930 Paris Salon .</code> | <code>1</code> |
247
- * Loss: [<code>MultipleNegativesSymmetricRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativessymmetricrankingloss) with these parameters:
248
  ```json
249
  {
250
  "scale": 20.0,
251
- "similarity_fct": "cos_sim",
252
- "gather_across_devices": false
253
  }
254
  ```
255
 
@@ -271,366 +270,19 @@ You can finetune this model on your own dataset.
271
  | <code>The newer Punts are still very much in existence today and race in the same fleets as the older boats .</code> | <code>The newer punts are still very much in existence today and run in the same fleets as the older boats .</code> | <code>1</code> |
272
  | <code>After losing his second election , he resigned as opposition leader and was replaced by Geoff Pearsall .</code> | <code>Max Bingham resigned as opposition leader after losing his second election , and was replaced by Geoff Pearsall .</code> | <code>1</code> |
273
  | <code>The 12F was officially homologated on August 21 , 1929 and exhibited at the Paris Salon in 1930 .</code> | <code>The 12F was officially homologated on 21 August 1929 and displayed at the 1930 Paris Salon .</code> | <code>1</code> |
274
- * Loss: [<code>MultipleNegativesSymmetricRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativessymmetricrankingloss) with these parameters:
275
  ```json
276
  {
277
  "scale": 20.0,
278
- "similarity_fct": "cos_sim",
279
- "gather_across_devices": false
280
  }
281
  ```
282
 
283
- ### Training Hyperparameters
284
- #### Non-Default Hyperparameters
285
-
286
- - `eval_strategy`: steps
287
- - `per_device_train_batch_size`: 256
288
- - `per_device_eval_batch_size`: 256
289
- - `learning_rate`: 0.0003
290
- - `adam_beta2`: 0.98
291
- - `adam_epsilon`: 1e-06
292
- - `max_steps`: 200000
293
- - `warmup_steps`: 1000
294
- - `load_best_model_at_end`: True
295
- - `optim`: adamw_torch
296
- - `ddp_find_unused_parameters`: False
297
- - `push_to_hub`: True
298
- - `hub_model_id`: redis/langcache-embed-v3
299
- - `batch_sampler`: no_duplicates
300
-
301
- #### All Hyperparameters
302
- <details><summary>Click to expand</summary>
303
-
304
- - `overwrite_output_dir`: False
305
- - `do_predict`: False
306
- - `eval_strategy`: steps
307
- - `prediction_loss_only`: True
308
- - `per_device_train_batch_size`: 256
309
- - `per_device_eval_batch_size`: 256
310
- - `per_gpu_train_batch_size`: None
311
- - `per_gpu_eval_batch_size`: None
312
- - `gradient_accumulation_steps`: 1
313
- - `eval_accumulation_steps`: None
314
- - `torch_empty_cache_steps`: None
315
- - `learning_rate`: 0.0003
316
- - `weight_decay`: 0.0
317
- - `adam_beta1`: 0.9
318
- - `adam_beta2`: 0.98
319
- - `adam_epsilon`: 1e-06
320
- - `max_grad_norm`: 1.0
321
- - `num_train_epochs`: 3.0
322
- - `max_steps`: 200000
323
- - `lr_scheduler_type`: linear
324
- - `lr_scheduler_kwargs`: {}
325
- - `warmup_ratio`: 0.0
326
- - `warmup_steps`: 1000
327
- - `log_level`: passive
328
- - `log_level_replica`: warning
329
- - `log_on_each_node`: True
330
- - `logging_nan_inf_filter`: True
331
- - `save_safetensors`: True
332
- - `save_on_each_node`: False
333
- - `save_only_model`: False
334
- - `restore_callback_states_from_checkpoint`: False
335
- - `no_cuda`: False
336
- - `use_cpu`: False
337
- - `use_mps_device`: False
338
- - `seed`: 42
339
- - `data_seed`: None
340
- - `jit_mode_eval`: False
341
- - `use_ipex`: False
342
- - `bf16`: False
343
- - `fp16`: False
344
- - `fp16_opt_level`: O1
345
- - `half_precision_backend`: auto
346
- - `bf16_full_eval`: False
347
- - `fp16_full_eval`: False
348
- - `tf32`: None
349
- - `local_rank`: 0
350
- - `ddp_backend`: None
351
- - `tpu_num_cores`: None
352
- - `tpu_metrics_debug`: False
353
- - `debug`: []
354
- - `dataloader_drop_last`: False
355
- - `dataloader_num_workers`: 0
356
- - `dataloader_prefetch_factor`: None
357
- - `past_index`: -1
358
- - `disable_tqdm`: False
359
- - `remove_unused_columns`: True
360
- - `label_names`: None
361
- - `load_best_model_at_end`: True
362
- - `ignore_data_skip`: False
363
- - `fsdp`: []
364
- - `fsdp_min_num_params`: 0
365
- - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
366
- - `fsdp_transformer_layer_cls_to_wrap`: None
367
- - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
368
- - `parallelism_config`: None
369
- - `deepspeed`: None
370
- - `label_smoothing_factor`: 0.0
371
- - `optim`: adamw_torch
372
- - `optim_args`: None
373
- - `adafactor`: False
374
- - `group_by_length`: False
375
- - `length_column_name`: length
376
- - `ddp_find_unused_parameters`: False
377
- - `ddp_bucket_cap_mb`: None
378
- - `ddp_broadcast_buffers`: False
379
- - `dataloader_pin_memory`: True
380
- - `dataloader_persistent_workers`: False
381
- - `skip_memory_metrics`: True
382
- - `use_legacy_prediction_loop`: False
383
- - `push_to_hub`: True
384
- - `resume_from_checkpoint`: None
385
- - `hub_model_id`: redis/langcache-embed-v3
386
- - `hub_strategy`: every_save
387
- - `hub_private_repo`: None
388
- - `hub_always_push`: False
389
- - `hub_revision`: None
390
- - `gradient_checkpointing`: False
391
- - `gradient_checkpointing_kwargs`: None
392
- - `include_inputs_for_metrics`: False
393
- - `include_for_metrics`: []
394
- - `eval_do_concat_batches`: True
395
- - `fp16_backend`: auto
396
- - `push_to_hub_model_id`: None
397
- - `push_to_hub_organization`: None
398
- - `mp_parameters`:
399
- - `auto_find_batch_size`: False
400
- - `full_determinism`: False
401
- - `torchdynamo`: None
402
- - `ray_scope`: last
403
- - `ddp_timeout`: 1800
404
- - `torch_compile`: False
405
- - `torch_compile_backend`: None
406
- - `torch_compile_mode`: None
407
- - `include_tokens_per_second`: False
408
- - `include_num_input_tokens_seen`: False
409
- - `neftune_noise_alpha`: None
410
- - `optim_target_modules`: None
411
- - `batch_eval_metrics`: False
412
- - `eval_on_start`: False
413
- - `use_liger_kernel`: False
414
- - `liger_kernel_config`: None
415
- - `eval_use_gather_object`: False
416
- - `average_tokens_across_devices`: False
417
- - `prompts`: None
418
- - `batch_sampler`: no_duplicates
419
- - `multi_dataset_batch_sampler`: proportional
420
- - `router_mapping`: {}
421
- - `learning_rate_mapping`: {}
422
-
423
- </details>
424
-
425
  ### Training Logs
426
- <details><summary>Click to expand</summary>
 
 
427
 
428
- | Epoch | Step | Training Loss | Validation Loss | train_cosine_ndcg@10 |
429
- |:-----------:|:---------:|:-------------:|:---------------:|:--------------------:|
430
- | -1 | -1 | - | - | 0.7522 |
431
- | 0.5291 | 1000 | 0.0231 | 0.1710 | 0.7518 |
432
- | 1.0582 | 2000 | 0.0147 | 0.1552 | 0.7593 |
433
- | 1.5873 | 3000 | 0.0126 | 0.1616 | 0.7603 |
434
- | 2.1164 | 4000 | 0.0113 | 0.1301 | 0.7644 |
435
- | 2.6455 | 5000 | 0.0119 | 0.1276 | 0.7659 |
436
- | 3.1746 | 6000 | 0.0099 | 0.1270 | 0.7648 |
437
- | 3.7037 | 7000 | 0.0101 | 0.1239 | 0.7676 |
438
- | 4.2328 | 8000 | 0.0093 | 0.1267 | 0.7709 |
439
- | 4.7619 | 9000 | 0.0092 | 0.1190 | 0.7711 |
440
- | 5.2910 | 10000 | 0.0088 | 0.1145 | 0.7735 |
441
- | 5.8201 | 11000 | 0.009 | 0.1172 | 0.7735 |
442
- | 6.3492 | 12000 | 0.0083 | 0.1144 | 0.7749 |
443
- | 6.8783 | 13000 | 0.0088 | 0.1140 | 0.7736 |
444
- | 7.4074 | 14000 | 0.0083 | 0.1134 | 0.7751 |
445
- | 7.9365 | 15000 | 0.0087 | 0.1108 | 0.7742 |
446
- | 8.4656 | 16000 | 0.0084 | 0.1119 | 0.7759 |
447
- | 8.9947 | 17000 | 0.0081 | 0.1125 | 0.7762 |
448
- | 9.5238 | 18000 | 0.0081 | 0.1134 | 0.7768 |
449
- | 10.0529 | 19000 | 0.008 | 0.1126 | 0.7766 |
450
- | 10.5820 | 20000 | 0.0079 | 0.1119 | 0.7755 |
451
- | 11.1111 | 21000 | 0.0078 | 0.1112 | 0.7781 |
452
- | 11.6402 | 22000 | 0.008 | 0.1113 | 0.7778 |
453
- | 12.1693 | 23000 | 0.0082 | 0.1066 | 0.7796 |
454
- | 12.6984 | 24000 | 0.0078 | 0.1098 | 0.7775 |
455
- | 13.2275 | 25000 | 0.0078 | 0.1089 | 0.7800 |
456
- | 13.7566 | 26000 | 0.0074 | 0.1091 | 0.7779 |
457
- | 14.2857 | 27000 | 0.0078 | 0.1061 | 0.7782 |
458
- | 14.8148 | 28000 | 0.0074 | 0.1073 | 0.7769 |
459
- | 15.3439 | 29000 | 0.0078 | 0.1022 | 0.7804 |
460
- | 15.8730 | 30000 | 0.0078 | 0.1035 | 0.7799 |
461
- | 16.4021 | 31000 | 0.0074 | 0.1046 | 0.7793 |
462
- | 16.9312 | 32000 | 0.0074 | 0.1043 | 0.7817 |
463
- | 17.4603 | 33000 | 0.0071 | 0.1056 | 0.7831 |
464
- | 17.9894 | 34000 | 0.0074 | 0.1022 | 0.7820 |
465
- | 18.5185 | 35000 | 0.0073 | 0.1035 | 0.7820 |
466
- | 19.0476 | 36000 | 0.0074 | 0.1020 | 0.7836 |
467
- | 19.5767 | 37000 | 0.0071 | 0.1036 | 0.7828 |
468
- | 20.1058 | 38000 | 0.007 | 0.1029 | 0.7845 |
469
- | 20.6349 | 39000 | 0.0071 | 0.1019 | 0.7835 |
470
- | 21.1640 | 40000 | 0.007 | 0.0991 | 0.7849 |
471
- | 21.6931 | 41000 | 0.0071 | 0.1013 | 0.7828 |
472
- | 22.2222 | 42000 | 0.0073 | 0.1033 | 0.7833 |
473
- | 22.7513 | 43000 | 0.0068 | 0.0996 | 0.7835 |
474
- | 23.2804 | 44000 | 0.007 | 0.0976 | 0.7850 |
475
- | 23.8095 | 45000 | 0.0069 | 0.0986 | 0.7840 |
476
- | 24.3386 | 46000 | 0.0068 | 0.0992 | 0.7856 |
477
- | 24.8677 | 47000 | 0.0068 | 0.0988 | 0.7838 |
478
- | 25.3968 | 48000 | 0.0068 | 0.0980 | 0.7857 |
479
- | 25.9259 | 49000 | 0.007 | 0.0976 | 0.7860 |
480
- | 26.4550 | 50000 | 0.0071 | 0.0994 | 0.7850 |
481
- | 26.9841 | 51000 | 0.0067 | 0.0984 | 0.7862 |
482
- | 27.5132 | 52000 | 0.0064 | 0.0992 | 0.7845 |
483
- | 28.0423 | 53000 | 0.0068 | 0.1021 | 0.7840 |
484
- | 28.5714 | 54000 | 0.0066 | 0.0974 | 0.7863 |
485
- | 29.1005 | 55000 | 0.0066 | 0.1001 | 0.7848 |
486
- | 29.6296 | 56000 | 0.0067 | 0.0997 | 0.7848 |
487
- | 30.1587 | 57000 | 0.0067 | 0.0965 | 0.7868 |
488
- | 30.6878 | 58000 | 0.0067 | 0.0968 | 0.7858 |
489
- | 31.2169 | 59000 | 0.0066 | 0.0973 | 0.7861 |
490
- | 31.7460 | 60000 | 0.0067 | 0.0972 | 0.7865 |
491
- | 32.2751 | 61000 | 0.0065 | 0.0991 | 0.7855 |
492
- | 32.8042 | 62000 | 0.0062 | 0.0960 | 0.7871 |
493
- | 33.3333 | 63000 | 0.0068 | 0.1006 | 0.7863 |
494
- | 33.8624 | 64000 | 0.0063 | 0.0980 | 0.7872 |
495
- | 34.3915 | 65000 | 0.0066 | 0.0957 | 0.7871 |
496
- | 34.9206 | 66000 | 0.0066 | 0.0971 | 0.7870 |
497
- | 35.4497 | 67000 | 0.0063 | 0.0982 | 0.7857 |
498
- | 35.9788 | 68000 | 0.0067 | 0.0944 | 0.7871 |
499
- | 36.5079 | 69000 | 0.0062 | 0.0961 | 0.7870 |
500
- | 37.0370 | 70000 | 0.0061 | 0.0924 | 0.7880 |
501
- | 37.5661 | 71000 | 0.0064 | 0.0928 | 0.7878 |
502
- | 38.0952 | 72000 | 0.0065 | 0.0934 | 0.7888 |
503
- | 38.6243 | 73000 | 0.0069 | 0.0948 | 0.7873 |
504
- | **39.1534** | **74000** | **0.0064** | **0.0922** | **0.7885** |
505
- | 39.6825 | 75000 | 0.0064 | 0.0937 | 0.7888 |
506
- | 40.2116 | 76000 | 0.0059 | 0.0941 | 0.7882 |
507
- | 40.7407 | 77000 | 0.0067 | 0.0934 | 0.7900 |
508
- | 41.2698 | 78000 | 0.0064 | 0.0926 | 0.7888 |
509
- | 41.7989 | 79000 | 0.006 | 0.0948 | 0.7880 |
510
- | 42.3280 | 80000 | 0.006 | 0.0953 | 0.7876 |
511
- | 42.8571 | 81000 | 0.0058 | 0.0955 | 0.7887 |
512
- | 43.3862 | 82000 | 0.0065 | 0.0945 | 0.7875 |
513
- | 43.9153 | 83000 | 0.0063 | 0.0928 | 0.7888 |
514
- | 44.4444 | 84000 | 0.0065 | 0.0959 | 0.7883 |
515
- | 44.9735 | 85000 | 0.0063 | 0.0956 | 0.7876 |
516
- | 45.5026 | 86000 | 0.006 | 0.0946 | 0.7893 |
517
- | 46.0317 | 87000 | 0.0062 | 0.0954 | 0.7908 |
518
- | 46.5608 | 88000 | 0.0061 | 0.0960 | 0.7896 |
519
- | 47.0899 | 89000 | 0.006 | 0.0953 | 0.7893 |
520
- | 47.6190 | 90000 | 0.0058 | 0.0941 | 0.7899 |
521
- | 48.1481 | 91000 | 0.0059 | 0.0950 | 0.7892 |
522
- | 48.6772 | 92000 | 0.0066 | 0.0948 | 0.7890 |
523
- | 49.2063 | 93000 | 0.0058 | 0.0947 | 0.7886 |
524
- | 49.7354 | 94000 | 0.006 | 0.0952 | 0.7891 |
525
- | 50.2646 | 95000 | 0.0058 | 0.0948 | 0.7885 |
526
- | 50.7937 | 96000 | 0.0058 | 0.0945 | 0.7894 |
527
- | 51.3228 | 97000 | 0.0059 | 0.0936 | 0.7901 |
528
- | 51.8519 | 98000 | 0.0059 | 0.0950 | 0.7900 |
529
- | 52.3810 | 99000 | 0.0058 | 0.0954 | 0.7893 |
530
- | 52.9101 | 100000 | 0.0058 | 0.0946 | 0.7900 |
531
- | 53.4392 | 101000 | 0.0056 | 0.0943 | 0.7900 |
532
- | 53.9683 | 102000 | 0.006 | 0.0950 | 0.7895 |
533
- | 54.4974 | 103000 | 0.0059 | 0.0937 | 0.7899 |
534
- | 55.0265 | 104000 | 0.0061 | 0.0941 | 0.7897 |
535
- | 55.5556 | 105000 | 0.0059 | 0.0941 | 0.7903 |
536
- | 56.0847 | 106000 | 0.0057 | 0.0924 | 0.7904 |
537
- | 56.6138 | 107000 | 0.006 | 0.0933 | 0.7901 |
538
- | 57.1429 | 108000 | 0.0059 | 0.0948 | 0.7888 |
539
- | 57.6720 | 109000 | 0.0061 | 0.0938 | 0.7899 |
540
- | 58.2011 | 110000 | 0.0058 | 0.0942 | 0.7904 |
541
- | 58.7302 | 111000 | 0.0056 | 0.0943 | 0.7913 |
542
- | 59.2593 | 112000 | 0.0056 | 0.0949 | 0.7915 |
543
- | 59.7884 | 113000 | 0.0058 | 0.0947 | 0.7907 |
544
- | 60.3175 | 114000 | 0.0058 | 0.0939 | 0.7910 |
545
- | 60.8466 | 115000 | 0.0058 | 0.0942 | 0.7906 |
546
- | 61.3757 | 116000 | 0.0055 | 0.0933 | 0.7910 |
547
- | 61.9048 | 117000 | 0.0055 | 0.0936 | 0.7913 |
548
- | 62.4339 | 118000 | 0.0059 | 0.0937 | 0.7904 |
549
- | 62.9630 | 119000 | 0.0057 | 0.0943 | 0.7908 |
550
- | 63.4921 | 120000 | 0.0056 | 0.0934 | 0.7912 |
551
- | 64.0212 | 121000 | 0.0058 | 0.0936 | 0.7909 |
552
- | 64.5503 | 122000 | 0.0055 | 0.0942 | 0.7896 |
553
- | 65.0794 | 123000 | 0.0058 | 0.0939 | 0.7901 |
554
- | 65.6085 | 124000 | 0.0057 | 0.0936 | 0.7907 |
555
- | 66.1376 | 125000 | 0.0054 | 0.0951 | 0.7901 |
556
- | 66.6667 | 126000 | 0.0055 | 0.0942 | 0.7912 |
557
- | 67.1958 | 127000 | 0.0057 | 0.0943 | 0.7914 |
558
- | 67.7249 | 128000 | 0.0057 | 0.0937 | 0.7910 |
559
- | 68.2540 | 129000 | 0.0057 | 0.0933 | 0.7918 |
560
- | 68.7831 | 130000 | 0.0055 | 0.0935 | 0.7913 |
561
- | 69.3122 | 131000 | 0.0053 | 0.0935 | 0.7908 |
562
- | 69.8413 | 132000 | 0.0057 | 0.0937 | 0.7905 |
563
- | 70.3704 | 133000 | 0.0055 | 0.0940 | 0.7912 |
564
- | 70.8995 | 134000 | 0.0052 | 0.0937 | 0.7913 |
565
- | 71.4286 | 135000 | 0.005 | 0.0940 | 0.7917 |
566
- | 71.9577 | 136000 | 0.0053 | 0.0933 | 0.7914 |
567
- | 72.4868 | 137000 | 0.0056 | 0.0940 | 0.7915 |
568
- | 73.0159 | 138000 | 0.0054 | 0.0937 | 0.7909 |
569
- | 73.5450 | 139000 | 0.0051 | 0.0940 | 0.7909 |
570
- | 74.0741 | 140000 | 0.0058 | 0.0938 | 0.7911 |
571
- | 74.6032 | 141000 | 0.0056 | 0.0938 | 0.7912 |
572
- | 75.1323 | 142000 | 0.0052 | 0.0931 | 0.7908 |
573
- | 75.6614 | 143000 | 0.0052 | 0.0937 | 0.7905 |
574
- | 76.1905 | 144000 | 0.0054 | 0.0940 | 0.7905 |
575
- | 76.7196 | 145000 | 0.0055 | 0.0940 | 0.7907 |
576
- | 77.2487 | 146000 | 0.0053 | 0.0941 | 0.7909 |
577
- | 77.7778 | 147000 | 0.0057 | 0.0944 | 0.7907 |
578
- | 78.3069 | 148000 | 0.0054 | 0.0947 | 0.7909 |
579
- | 78.8360 | 149000 | 0.0054 | 0.0949 | 0.7907 |
580
- | 79.3651 | 150000 | 0.0055 | 0.0948 | 0.7907 |
581
- | 79.8942 | 151000 | 0.0058 | 0.0950 | 0.7907 |
582
- | 80.4233 | 152000 | 0.0054 | 0.0946 | 0.7907 |
583
- | 80.9524 | 153000 | 0.0053 | 0.0949 | 0.7909 |
584
- | 81.4815 | 154000 | 0.0055 | 0.0947 | 0.7908 |
585
- | 82.0106 | 155000 | 0.0053 | 0.0946 | 0.7906 |
586
- | 82.5397 | 156000 | 0.0053 | 0.0949 | 0.7906 |
587
- | 83.0688 | 157000 | 0.0051 | 0.0948 | 0.7912 |
588
- | 83.5979 | 158000 | 0.0052 | 0.0954 | 0.7906 |
589
- | 84.1270 | 159000 | 0.0054 | 0.0953 | 0.7908 |
590
- | 84.6561 | 160000 | 0.005 | 0.0951 | 0.7911 |
591
- | 85.1852 | 161000 | 0.0054 | 0.0953 | 0.7910 |
592
- | 85.7143 | 162000 | 0.0056 | 0.0957 | 0.7907 |
593
- | 86.2434 | 163000 | 0.0054 | 0.0953 | 0.7909 |
594
- | 86.7725 | 164000 | 0.0051 | 0.0955 | 0.7912 |
595
- | 87.3016 | 165000 | 0.0055 | 0.0956 | 0.7911 |
596
- | 87.8307 | 166000 | 0.0056 | 0.0954 | 0.7909 |
597
- | 88.3598 | 167000 | 0.0052 | 0.0955 | 0.7911 |
598
- | 88.8889 | 168000 | 0.0052 | 0.0953 | 0.7910 |
599
- | 89.4180 | 169000 | 0.0052 | 0.0952 | 0.7906 |
600
- | 89.9471 | 170000 | 0.0053 | 0.0952 | 0.7908 |
601
- | 90.4762 | 171000 | 0.0052 | 0.0954 | 0.7908 |
602
- | 91.0053 | 172000 | 0.0054 | 0.0954 | 0.7907 |
603
- | 91.5344 | 173000 | 0.0052 | 0.0951 | 0.7909 |
604
- | 92.0635 | 174000 | 0.0053 | 0.0951 | 0.7907 |
605
- | 92.5926 | 175000 | 0.0051 | 0.0950 | 0.7906 |
606
- | 93.1217 | 176000 | 0.0054 | 0.0953 | 0.7907 |
607
- | 93.6508 | 177000 | 0.0052 | 0.0953 | 0.7907 |
608
- | 94.1799 | 178000 | 0.0051 | 0.0951 | 0.7908 |
609
- | 94.7090 | 179000 | 0.0052 | 0.0952 | 0.7906 |
610
- | 95.2381 | 180000 | 0.0053 | 0.0953 | 0.7909 |
611
- | 95.7672 | 181000 | 0.0052 | 0.0953 | 0.7908 |
612
- | 96.2963 | 182000 | 0.0051 | 0.0952 | 0.7906 |
613
- | 96.8254 | 183000 | 0.0053 | 0.0953 | 0.7907 |
614
- | 97.3545 | 184000 | 0.0051 | 0.0953 | 0.7907 |
615
- | 97.8836 | 185000 | 0.0049 | 0.0953 | 0.7906 |
616
- | 98.4127 | 186000 | 0.0051 | 0.0953 | 0.7907 |
617
- | 98.9418 | 187000 | 0.0051 | 0.0954 | 0.7906 |
618
- | 99.4709 | 188000 | 0.0053 | 0.0954 | 0.7906 |
619
- | 100.0 | 189000 | 0.0051 | 0.0954 | 0.7904 |
620
- | 100.5291 | 190000 | 0.0054 | 0.0953 | 0.7907 |
621
- | 101.0582 | 191000 | 0.0052 | 0.0954 | 0.7905 |
622
- | 101.5873 | 192000 | 0.0051 | 0.0954 | 0.7907 |
623
- | 102.1164 | 193000 | 0.0052 | 0.0953 | 0.7907 |
624
- | 102.6455 | 194000 | 0.0051 | 0.0955 | 0.7908 |
625
- | 103.1746 | 195000 | 0.0054 | 0.0954 | 0.7906 |
626
- | 103.7037 | 196000 | 0.0052 | 0.0954 | 0.7905 |
627
- | 104.2328 | 197000 | 0.0053 | 0.0954 | 0.7906 |
628
- | 104.7619 | 198000 | 0.0052 | 0.0954 | 0.7907 |
629
- | 105.2910 | 199000 | 0.0055 | 0.0954 | 0.7904 |
630
- | 105.8201 | 200000 | 0.0054 | 0.0955 | 0.7905 |
631
-
632
- * The bold row denotes the saved checkpoint.
633
- </details>
634
 
635
  ### Framework Versions
636
  - Python: 3.12.3
@@ -658,6 +310,17 @@ You can finetune this model on your own dataset.
658
  }
659
  ```
660
 
 
 
 
 
 
 
 
 
 
 
 
661
  <!--
662
  ## Glossary
663
 
 
12
  - retrieval
13
  - reranking
14
  - generated_from_trainer
15
+ - dataset_size:1056095
16
+ - loss:CoSENTLoss
17
  base_model: Alibaba-NLP/gte-modernbert-base
18
  widget:
19
  - source_sentence: 'See Precambrian time scale # Proposed Geologic timeline for another
 
85
  type: train
86
  metrics:
87
  - type: cosine_accuracy@1
88
+ value: 0.5579129681749296
89
  name: Cosine Accuracy@1
90
  - type: cosine_precision@1
91
+ value: 0.5579129681749296
92
  name: Cosine Precision@1
93
  - type: cosine_recall@1
94
+ value: 0.5359784831006956
95
  name: Cosine Recall@1
96
  - type: cosine_ndcg@10
97
+ value: 0.7522148521266401
98
  name: Cosine Ndcg@10
99
  - type: cosine_mrr@1
100
+ value: 0.5579129681749296
101
  name: Cosine Mrr@1
102
  - type: cosine_map@100
103
+ value: 0.6974638651409195
104
  name: Cosine Map@100
105
  ---
106
 
 
165
  # Get the similarity scores for the embeddings
166
  similarities = model.similarity(embeddings, embeddings)
167
  print(similarities)
168
+ # tensor([[0.9922, 0.9922, 0.5352],
169
+ # [0.9922, 0.9961, 0.5391],
170
+ # [0.5352, 0.5391, 1.0000]], dtype=torch.bfloat16)
171
  ```
172
 
173
  <!--
 
205
 
206
  | Metric | Value |
207
  |:-------------------|:-----------|
208
+ | cosine_accuracy@1 | 0.5579 |
209
+ | cosine_precision@1 | 0.5579 |
210
+ | cosine_recall@1 | 0.536 |
211
+ | **cosine_ndcg@10** | **0.7522** |
212
+ | cosine_mrr@1 | 0.5579 |
213
+ | cosine_map@100 | 0.6975 |
214
 
215
  <!--
216
  ## Bias, Risks and Limitations
 
244
  | <code>The newer Punts are still very much in existence today and race in the same fleets as the older boats .</code> | <code>The newer punts are still very much in existence today and run in the same fleets as the older boats .</code> | <code>1</code> |
245
  | <code>After losing his second election , he resigned as opposition leader and was replaced by Geoff Pearsall .</code> | <code>Max Bingham resigned as opposition leader after losing his second election , and was replaced by Geoff Pearsall .</code> | <code>1</code> |
246
  | <code>The 12F was officially homologated on August 21 , 1929 and exhibited at the Paris Salon in 1930 .</code> | <code>The 12F was officially homologated on 21 August 1929 and displayed at the 1930 Paris Salon .</code> | <code>1</code> |
247
+ * Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
248
  ```json
249
  {
250
  "scale": 20.0,
251
+ "similarity_fct": "pairwise_cos_sim"
 
252
  }
253
  ```
254
 
 
270
  | <code>The newer Punts are still very much in existence today and race in the same fleets as the older boats .</code> | <code>The newer punts are still very much in existence today and run in the same fleets as the older boats .</code> | <code>1</code> |
271
  | <code>After losing his second election , he resigned as opposition leader and was replaced by Geoff Pearsall .</code> | <code>Max Bingham resigned as opposition leader after losing his second election , and was replaced by Geoff Pearsall .</code> | <code>1</code> |
272
  | <code>The 12F was officially homologated on August 21 , 1929 and exhibited at the Paris Salon in 1930 .</code> | <code>The 12F was officially homologated on 21 August 1929 and displayed at the 1930 Paris Salon .</code> | <code>1</code> |
273
+ * Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
274
  ```json
275
  {
276
  "scale": 20.0,
277
+ "similarity_fct": "pairwise_cos_sim"
 
278
  }
279
  ```
280
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
281
  ### Training Logs
282
+ | Epoch | Step | train_cosine_ndcg@10 |
283
+ |:-----:|:----:|:--------------------:|
284
+ | -1 | -1 | 0.7522 |
285
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
286
 
287
  ### Framework Versions
288
  - Python: 3.12.3
 
310
  }
311
  ```
312
 
313
+ #### CoSENTLoss
314
+ ```bibtex
315
+ @online{kexuefm-8847,
316
+ title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
317
+ author={Su Jianlin},
318
+ year={2022},
319
+ month={Jan},
320
+ url={https://kexue.fm/archives/8847},
321
+ }
322
+ ```
323
+
324
  <!--
325
  ## Glossary
326
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e237a4c2a7b79b02474b49db1e9adb01059b9eb2d67a4fc8dca8db043f776c1c
3
  size 298041696
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95d02211c4cca89113f9f3e93ed91f5176bf50170faa2cb835f7bfea15bb9dd2
3
  size 298041696