Training checkpoint - Epoch 3, Step 33844

checkpoint-33844/1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 384,
+    "pooling_mode_cls_token": false,
+    "pooling_mode_mean_tokens": true,
+    "pooling_mode_max_tokens": false,
+    "pooling_mode_mean_sqrt_len_tokens": false,
+    "pooling_mode_weightedmean_tokens": false,
+    "pooling_mode_lasttoken": false,
+    "include_prompt": true
+}

checkpoint-33844/config.json

@@ -0,0 +1,25 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.53.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

checkpoint-33844/config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "__version__": {
+    "sentence_transformers": "5.2.0",
+    "transformers": "4.53.3",
+    "pytorch": "2.6.0+cu124"
+  },
+  "model_type": "SentenceTransformer",
+  "prompts": {
+    "query": "",
+    "document": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

checkpoint-33844/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33915727eca86abc0e88c39b626b53a81006e68383aab5a8e90cdcd0cd667660
+size 90864192

checkpoint-33844/modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

checkpoint-33844/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:261ce9bf5f0e0a23d905416f278b392fbff9afa5f3540c95e2203dda71c420f1
+size 180607738

checkpoint-33844/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f70e5d3990cf34650ffafdf1b4cbb5569975f5d3dc5811740d1fdb5ac5ae164c
+size 14244

checkpoint-33844/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17f82ce88c146b499dc628e4d82bb6f394081aeac6786996882944b1fb69de13
+size 988

checkpoint-33844/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2c86a52276791d7faf33b102232654822f00369861a4acaaa15cff0f770cb29
+size 1064

checkpoint-33844/sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 256,
+    "do_lower_case": false
+}

checkpoint-33844/special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "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": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "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
+  }
+}

checkpoint-33844/tokenizer_config.json

@@ -0,0 +1,65 @@
+{
+  "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
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "max_length": 128,
+  "model_max_length": 256,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}

checkpoint-33844/trainer_state.json

@@ -0,0 +1,297 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.9971661353170385,
+  "eval_steps": 2000,
+  "global_step": 33844,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00011818933932159319,
+      "grad_norm": 7.239190101623535,
+      "learning_rate": 0.0,
+      "loss": 3.9597,
+      "step": 1
+    },
+    {
+      "epoch": 0.23637867864318637,
+      "grad_norm": 4.569056034088135,
+      "learning_rate": 7.088652482269504e-06,
+      "loss": 3.3526,
+      "step": 2000
+    },
+    {
+      "epoch": 0.23637867864318637,
+      "eval_cosine_accuracy": 0.9351803064346313,
+      "eval_loss": 1.1521742343902588,
+      "eval_runtime": 22.8749,
+      "eval_samples_per_second": 413.423,
+      "eval_steps_per_second": 1.617,
+      "step": 2000
+    },
+    {
+      "epoch": 0.47275735728637275,
+      "grad_norm": 4.866804599761963,
+      "learning_rate": 1.4180851063829787e-05,
+      "loss": 2.6297,
+      "step": 4000
+    },
+    {
+      "epoch": 0.47275735728637275,
+      "eval_cosine_accuracy": 0.9370836615562439,
+      "eval_loss": 1.1368498802185059,
+      "eval_runtime": 24.2219,
+      "eval_samples_per_second": 390.433,
+      "eval_steps_per_second": 1.528,
+      "step": 4000
+    },
+    {
+      "epoch": 0.7091360359295592,
+      "grad_norm": 4.78344202041626,
+      "learning_rate": 2.126950354609929e-05,
+      "loss": 2.27,
+      "step": 6000
+    },
+    {
+      "epoch": 0.7091360359295592,
+      "eval_cosine_accuracy": 0.9424764513969421,
+      "eval_loss": 1.088667392730713,
+      "eval_runtime": 23.9037,
+      "eval_samples_per_second": 395.63,
+      "eval_steps_per_second": 1.548,
+      "step": 6000
+    },
+    {
+      "epoch": 0.9455147145727455,
+      "grad_norm": 4.2486572265625,
+      "learning_rate": 2.8358156028368792e-05,
+      "loss": 2.0264,
+      "step": 8000
+    },
+    {
+      "epoch": 0.9455147145727455,
+      "eval_cosine_accuracy": 0.9423707127571106,
+      "eval_loss": 1.0786563158035278,
+      "eval_runtime": 23.6661,
+      "eval_samples_per_second": 399.601,
+      "eval_steps_per_second": 1.563,
+      "step": 8000
+    },
+    {
+      "epoch": 1.181721572794899,
+      "grad_norm": 3.9295907020568848,
+      "learning_rate": 2.9729781514064122e-05,
+      "loss": 1.8456,
+      "step": 10000
+    },
+    {
+      "epoch": 1.181721572794899,
+      "eval_cosine_accuracy": 0.9449085593223572,
+      "eval_loss": 1.0610214471817017,
+      "eval_runtime": 23.723,
+      "eval_samples_per_second": 398.643,
+      "eval_steps_per_second": 1.56,
+      "step": 10000
+    },
+    {
+      "epoch": 1.4178769630416814,
+      "grad_norm": 4.169382572174072,
+      "learning_rate": 2.8587197732951956e-05,
+      "loss": 1.7248,
+      "step": 12000
+    },
+    {
+      "epoch": 1.4178769630416814,
+      "eval_cosine_accuracy": 0.9460716843605042,
+      "eval_loss": 1.0575031042099,
+      "eval_runtime": 23.9067,
+      "eval_samples_per_second": 395.579,
+      "eval_steps_per_second": 1.548,
+      "step": 12000
+    },
+    {
+      "epoch": 1.6540323532884638,
+      "grad_norm": 4.164876937866211,
+      "learning_rate": 2.6616033219669403e-05,
+      "loss": 1.6326,
+      "step": 14000
+    },
+    {
+      "epoch": 1.6540323532884638,
+      "eval_cosine_accuracy": 0.9483979940414429,
+      "eval_loss": 1.0424392223358154,
+      "eval_runtime": 23.6735,
+      "eval_samples_per_second": 399.476,
+      "eval_steps_per_second": 1.563,
+      "step": 14000
+    },
+    {
+      "epoch": 1.8901877435352463,
+      "grad_norm": 3.8136980533599854,
+      "learning_rate": 2.393977593462922e-05,
+      "loss": 1.5739,
+      "step": 16000
+    },
+    {
+      "epoch": 1.8901877435352463,
+      "eval_cosine_accuracy": 0.9505128264427185,
+      "eval_loss": 1.003242015838623,
+      "eval_runtime": 24.0212,
+      "eval_samples_per_second": 393.694,
+      "eval_steps_per_second": 1.54,
+      "step": 16000
+    },
+    {
+      "epoch": 2.1263431337820284,
+      "grad_norm": 3.914799690246582,
+      "learning_rate": 2.071794740754529e-05,
+      "loss": 1.5104,
+      "step": 18000
+    },
+    {
+      "epoch": 2.1263431337820284,
+      "eval_cosine_accuracy": 0.9495611786842346,
+      "eval_loss": 1.01349675655365,
+      "eval_runtime": 26.418,
+      "eval_samples_per_second": 357.976,
+      "eval_steps_per_second": 1.401,
+      "step": 18000
+    },
+    {
+      "epoch": 2.362498524028811,
+      "grad_norm": 3.88228440284729,
+      "learning_rate": 1.7146081315912403e-05,
+      "loss": 1.4683,
+      "step": 20000
+    },
+    {
+      "epoch": 2.362498524028811,
+      "eval_cosine_accuracy": 0.9518874883651733,
+      "eval_loss": 0.9978452324867249,
+      "eval_runtime": 24.04,
+      "eval_samples_per_second": 393.387,
+      "eval_steps_per_second": 1.539,
+      "step": 20000
+    },
+    {
+      "epoch": 2.5986539142755936,
+      "grad_norm": 3.965975761413574,
+      "learning_rate": 1.3445243725964677e-05,
+      "loss": 1.4373,
+      "step": 22000
+    },
+    {
+      "epoch": 2.5986539142755936,
+      "eval_cosine_accuracy": 0.954002320766449,
+      "eval_loss": 0.9861479997634888,
+      "eval_runtime": 40.0431,
+      "eval_samples_per_second": 236.17,
+      "eval_steps_per_second": 0.924,
+      "step": 22000
+    },
+    {
+      "epoch": 2.8348093045223757,
+      "grad_norm": 3.817960500717163,
+      "learning_rate": 9.839084604957456e-06,
+      "loss": 1.4122,
+      "step": 24000
+    },
+    {
+      "epoch": 2.8348093045223757,
+      "eval_cosine_accuracy": 0.9523104429244995,
+      "eval_loss": 0.9923800230026245,
+      "eval_runtime": 25.2947,
+      "eval_samples_per_second": 373.873,
+      "eval_steps_per_second": 1.463,
+      "step": 24000
+    },
+    {
+      "epoch": 3.0709646947691582,
+      "grad_norm": 3.5702011585235596,
+      "learning_rate": 6.547204690340503e-06,
+      "loss": 1.3984,
+      "step": 26000
+    },
+    {
+      "epoch": 3.0709646947691582,
+      "eval_cosine_accuracy": 0.9538965821266174,
+      "eval_loss": 0.9917270541191101,
+      "eval_runtime": 24.7317,
+      "eval_samples_per_second": 382.383,
+      "eval_steps_per_second": 1.496,
+      "step": 26000
+    },
+    {
+      "epoch": 3.3071200850159403,
+      "grad_norm": 4.189650058746338,
+      "learning_rate": 3.7688357224379343e-06,
+      "loss": 1.373,
+      "step": 28000
+    },
+    {
+      "epoch": 3.3071200850159403,
+      "eval_cosine_accuracy": 0.9547424912452698,
+      "eval_loss": 0.984880805015564,
+      "eval_runtime": 24.8474,
+      "eval_samples_per_second": 380.603,
+      "eval_steps_per_second": 1.489,
+      "step": 28000
+    },
+    {
+      "epoch": 3.543275475262723,
+      "grad_norm": 3.7039976119995117,
+      "learning_rate": 1.6767863047503395e-06,
+      "loss": 1.3683,
+      "step": 30000
+    },
+    {
+      "epoch": 3.543275475262723,
+      "eval_cosine_accuracy": 0.9554827213287354,
+      "eval_loss": 0.9861717224121094,
+      "eval_runtime": 25.3133,
+      "eval_samples_per_second": 373.598,
+      "eval_steps_per_second": 1.462,
+      "step": 30000
+    },
+    {
+      "epoch": 3.7794308655095055,
+      "grad_norm": 3.7472827434539795,
+      "learning_rate": 3.944139019812121e-07,
+      "loss": 1.361,
+      "step": 32000
+    },
+    {
+      "epoch": 3.7794308655095055,
+      "eval_cosine_accuracy": 0.9551655054092407,
+      "eval_loss": 0.9865333437919617,
+      "eval_runtime": 24.9113,
+      "eval_samples_per_second": 379.627,
+      "eval_steps_per_second": 1.485,
+      "step": 32000
+    }
+  ],
+  "logging_steps": 2000,
+  "max_steps": 33844,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-33844/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28ccac7a99a6ce97f3a532db37d53c85081bb52675bd1bcc884e2bb66487e67c
+size 5624

checkpoint-33844/training_metrics.json

@@ -0,0 +1,102 @@
+{
+  "metrics_history": [
+    {
+      "epoch": 0.23637867864318637,
+      "global_step": 2000,
+      "eval_loss": 1.1521742343902588,
+      "eval_cosine_accuracy": 0.9351803064346313
+    },
+    {
+      "epoch": 0.47275735728637275,
+      "global_step": 4000,
+      "eval_loss": 1.1368498802185059,
+      "eval_cosine_accuracy": 0.9370836615562439
+    },
+    {
+      "epoch": 0.7091360359295592,
+      "global_step": 6000,
+      "eval_loss": 1.088667392730713,
+      "eval_cosine_accuracy": 0.9424764513969421
+    },
+    {
+      "epoch": 0.9455147145727455,
+      "global_step": 8000,
+      "eval_loss": 1.0786563158035278,
+      "eval_cosine_accuracy": 0.9423707127571106
+    },
+    {
+      "epoch": 1.181721572794899,
+      "global_step": 10000,
+      "eval_loss": 1.0610214471817017,
+      "eval_cosine_accuracy": 0.9449085593223572
+    },
+    {
+      "epoch": 1.4178769630416814,
+      "global_step": 12000,
+      "eval_loss": 1.0575031042099,
+      "eval_cosine_accuracy": 0.9460716843605042
+    },
+    {
+      "epoch": 1.6540323532884638,
+      "global_step": 14000,
+      "eval_loss": 1.0424392223358154,
+      "eval_cosine_accuracy": 0.9483979940414429
+    },
+    {
+      "epoch": 1.8901877435352463,
+      "global_step": 16000,
+      "eval_loss": 1.003242015838623,
+      "eval_cosine_accuracy": 0.9505128264427185
+    },
+    {
+      "epoch": 2.1263431337820284,
+      "global_step": 18000,
+      "eval_loss": 1.01349675655365,
+      "eval_cosine_accuracy": 0.9495611786842346
+    },
+    {
+      "epoch": 2.362498524028811,
+      "global_step": 20000,
+      "eval_loss": 0.9978452324867249,
+      "eval_cosine_accuracy": 0.9518874883651733
+    },
+    {
+      "epoch": 2.5986539142755936,
+      "global_step": 22000,
+      "eval_loss": 0.9861479997634888,
+      "eval_cosine_accuracy": 0.954002320766449
+    },
+    {
+      "epoch": 2.8348093045223757,
+      "global_step": 24000,
+      "eval_loss": 0.9923800230026245,
+      "eval_cosine_accuracy": 0.9523104429244995
+    },
+    {
+      "epoch": 3.0709646947691582,
+      "global_step": 26000,
+      "eval_loss": 0.9917270541191101,
+      "eval_cosine_accuracy": 0.9538965821266174
+    },
+    {
+      "epoch": 3.3071200850159403,
+      "global_step": 28000,
+      "eval_loss": 0.984880805015564,
+      "eval_cosine_accuracy": 0.9547424912452698
+    },
+    {
+      "epoch": 3.543275475262723,
+      "global_step": 30000,
+      "eval_loss": 0.9861717224121094,
+      "eval_cosine_accuracy": 0.9554827213287354
+    },
+    {
+      "epoch": 3.7794308655095055,
+      "global_step": 32000,
+      "eval_loss": 0.9865333437919617,
+      "eval_cosine_accuracy": 0.9551655054092407
+    }
+  ],
+  "current_epoch": 3,
+  "global_step": 33844
+}