Hugging Face's logo

akhooli
/
sbert-nli-500k-triplets-MB

Sentence Similarity
sentence-transformers
TensorBoard
Safetensors
Arabic
modernbert
feature-extraction
Generated from Trainer
dataset_size:550000
loss:MatryoshkaLoss
loss:CachedMultipleNegativesRankingLoss
Inference Endpoints
Model card Files Metrics Community
akhooli commited on
Commit
73ca7f3
·
verified ·
1 Parent(s): d95bdf3

akhooli/nli-500k-triplets-MB

Browse files
Files changed (4) hide show
  1. README.md +92 -7
  2. model.safetensors +1 -1
  3. runs/Jan15_08-03-45_21ac67580ab7/events.out.tfevents.1736928228.21ac67580ab7.99.0 +3 -0
  4. training_args.bin +1 -1
README.md CHANGED
@@ -281,7 +281,6 @@ You can finetune this model on your own dataset.
281
  - `per_device_eval_batch_size`: 12
282
  - `learning_rate`: 2e-05
283
  - `num_train_epochs`: 1
284
- - `max_steps`: 500
285
  - `warmup_ratio`: 0.1
286
  - `fp16`: True
287
  - `batch_sampler`: no_duplicates
@@ -307,7 +306,7 @@ You can finetune this model on your own dataset.
307
  - `adam_epsilon`: 1e-08
308
  - `max_grad_norm`: 1.0
309
  - `num_train_epochs`: 1
310
- - `max_steps`: 500
311
  - `lr_scheduler_type`: linear
312
  - `lr_scheduler_kwargs`: {}
313
  - `warmup_ratio`: 0.1
@@ -408,10 +407,96 @@ You can finetune this model on your own dataset.
408
  </details>
409
 
410
  ### Training Logs
411
- | Epoch | Step | Training Loss |
412
- |:------:|:----:|:-------------:|
413
- | 0.0113 | 250 | 4.6054 |
414
- | 0.0226 | 500 | 3.3232 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
415
 
416
 
417
  ### Framework Versions
@@ -420,7 +505,7 @@ You can finetune this model on your own dataset.
420
  - Transformers: 4.48.0
421
  - PyTorch: 2.5.1+cu121
422
  - Accelerate: 0.34.2
423
- - Datasets: 3.0.0
424
  - Tokenizers: 0.21.0
425
 
426
  ## Citation
 
281
  - `per_device_eval_batch_size`: 12
282
  - `learning_rate`: 2e-05
283
  - `num_train_epochs`: 1
 
284
  - `warmup_ratio`: 0.1
285
  - `fp16`: True
286
  - `batch_sampler`: no_duplicates
 
306
  - `adam_epsilon`: 1e-08
307
  - `max_grad_norm`: 1.0
308
  - `num_train_epochs`: 1
309
+ - `max_steps`: -1
310
  - `lr_scheduler_type`: linear
311
  - `lr_scheduler_kwargs`: {}
312
  - `warmup_ratio`: 0.1
 
407
  </details>
408
 
409
  ### Training Logs
410
+ | Epoch | Step | Training Loss | Validation Loss |
411
+ |:------:|:-----:|:-------------:|:---------------:|
412
+ | 0.0113 | 250 | 6.9693 | - |
413
+ | 0.0226 | 500 | 4.6555 | - |
414
+ | 0.0340 | 750 | 3.6612 | - |
415
+ | 0.0453 | 1000 | 3.3137 | - |
416
+ | 0.0566 | 1250 | 3.0655 | - |
417
+ | 0.0679 | 1500 | 2.9465 | - |
418
+ | 0.0792 | 1750 | 2.9436 | - |
419
+ | 0.0906 | 2000 | 2.7902 | - |
420
+ | 0.1019 | 2250 | 2.7131 | - |
421
+ | 0.1132 | 2500 | 2.7388 | - |
422
+ | 0.1245 | 2750 | 2.7474 | - |
423
+ | 0.1359 | 3000 | 2.5196 | - |
424
+ | 0.1472 | 3250 | 2.4522 | - |
425
+ | 0.1585 | 3500 | 2.4588 | - |
426
+ | 0.1698 | 3750 | 2.458 | - |
427
+ | 0.1811 | 4000 | 2.3643 | - |
428
+ | 0.1925 | 4250 | 2.2747 | - |
429
+ | 0.2038 | 4500 | 2.1526 | - |
430
+ | 0.2151 | 4750 | 2.0504 | - |
431
+ | 0.2264 | 5000 | 2.1278 | 2.1682 |
432
+ | 0.2377 | 5250 | 2.0536 | - |
433
+ | 0.2491 | 5500 | 2.0332 | - |
434
+ | 0.2604 | 5750 | 1.9816 | - |
435
+ | 0.2717 | 6000 | 1.8878 | - |
436
+ | 0.2830 | 6250 | 1.8733 | - |
437
+ | 0.2943 | 6500 | 1.8573 | - |
438
+ | 0.3057 | 6750 | 1.9132 | - |
439
+ | 0.3170 | 7000 | 1.7868 | - |
440
+ | 0.3283 | 7250 | 1.7047 | - |
441
+ | 0.3396 | 7500 | 1.836 | - |
442
+ | 0.3509 | 7750 | 1.7552 | - |
443
+ | 0.3623 | 8000 | 1.6976 | - |
444
+ | 0.3736 | 8250 | 1.7005 | - |
445
+ | 0.3849 | 8500 | 1.7418 | - |
446
+ | 0.3962 | 8750 | 1.6407 | - |
447
+ | 0.4076 | 9000 | 1.6039 | - |
448
+ | 0.4189 | 9250 | 1.6287 | - |
449
+ | 0.4302 | 9500 | 1.5528 | - |
450
+ | 0.4415 | 9750 | 1.5981 | - |
451
+ | 0.4528 | 10000 | 1.5705 | 1.6057 |
452
+ | 0.4642 | 10250 | 1.5245 | - |
453
+ | 0.4755 | 10500 | 1.444 | - |
454
+ | 0.4868 | 10750 | 1.4127 | - |
455
+ | 0.4981 | 11000 | 1.5006 | - |
456
+ | 0.5094 | 11250 | 1.3875 | - |
457
+ | 0.5208 | 11500 | 1.3678 | - |
458
+ | 0.5321 | 11750 | 1.4748 | - |
459
+ | 0.5434 | 12000 | 1.4333 | - |
460
+ | 0.5547 | 12250 | 1.4464 | - |
461
+ | 0.5660 | 12500 | 1.3053 | - |
462
+ | 0.5774 | 12750 | 1.3522 | - |
463
+ | 0.5887 | 13000 | 1.2708 | - |
464
+ | 0.6000 | 13250 | 1.3063 | - |
465
+ | 0.6113 | 13500 | 1.329 | - |
466
+ | 0.6227 | 13750 | 1.2761 | - |
467
+ | 0.6340 | 14000 | 1.303 | - |
468
+ | 0.6453 | 14250 | 1.3323 | - |
469
+ | 0.6566 | 14500 | 1.2228 | - |
470
+ | 0.6679 | 14750 | 1.2488 | - |
471
+ | 0.6793 | 15000 | 1.2635 | 1.2501 |
472
+ | 0.6906 | 15250 | 1.2333 | - |
473
+ | 0.7019 | 15500 | 1.203 | - |
474
+ | 0.7132 | 15750 | 1.1859 | - |
475
+ | 0.7245 | 16000 | 1.166 | - |
476
+ | 0.7359 | 16250 | 1.1935 | - |
477
+ | 0.7472 | 16500 | 1.1466 | - |
478
+ | 0.7585 | 16750 | 1.2175 | - |
479
+ | 0.7698 | 17000 | 1.1073 | - |
480
+ | 0.7811 | 17250 | 1.1363 | - |
481
+ | 0.7925 | 17500 | 1.2153 | - |
482
+ | 0.8038 | 17750 | 1.1302 | - |
483
+ | 0.8151 | 18000 | 1.0426 | - |
484
+ | 0.8264 | 18250 | 1.0576 | - |
485
+ | 0.8377 | 18500 | 1.0944 | - |
486
+ | 0.8491 | 18750 | 1.0158 | - |
487
+ | 0.8604 | 19000 | 1.1024 | - |
488
+ | 0.8717 | 19250 | 1.0598 | - |
489
+ | 0.8830 | 19500 | 1.0636 | - |
490
+ | 0.8944 | 19750 | 1.0052 | - |
491
+ | 0.9057 | 20000 | 1.0509 | 1.0778 |
492
+ | 0.9170 | 20250 | 1.0107 | - |
493
+ | 0.9283 | 20500 | 0.9524 | - |
494
+ | 0.9396 | 20750 | 1.0021 | - |
495
+ | 0.9510 | 21000 | 1.0262 | - |
496
+ | 0.9623 | 21250 | 0.9597 | - |
497
+ | 0.9736 | 21500 | 0.9471 | - |
498
+ | 0.9849 | 21750 | 1.0467 | - |
499
+ | 0.9962 | 22000 | 1.051 | - |
500
 
501
 
502
  ### Framework Versions
 
505
  - Transformers: 4.48.0
506
  - PyTorch: 2.5.1+cu121
507
  - Accelerate: 0.34.2
508
+ - Datasets: 3.0.1
509
  - Tokenizers: 0.21.0
510
 
511
  ## Citation
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3f025a5718ff97f31a2eaa8bbcd4e4a6d77a613052cf0c733ab7a6926cd92825
3
  size 596070136
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f92d69b7df664f82568c3d7637f84d55d7070729d683e18d8b3b8f9e498156e
3
  size 596070136
runs/Jan15_08-03-45_21ac67580ab7/events.out.tfevents.1736928228.21ac67580ab7.99.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:260c123fa491a2820a1ff1d5a1266da89326ff120c3bda2836fe50ead3f1e400
3
+ size 24545
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fd9ecd4bb16d7d71bdb6f21a4710945dcf5fd8fade9734374135c36a05dfb4b3
3
  size 5688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10148b8160d80da588a514decb1807d3d9168d3a4a81000638093b3acb0983a6
3
  size 5688