Training in progress, epoch 8, checkpoint

checkpoint-648/config.json

@@ -0,0 +1,53 @@
+{
+  "_name_or_path": "allenai/longformer-base-4096",
+  "architectures": [
+    "LongformerForTokenClassification"
+  ],
+  "attention_mode": "longformer",
+  "attention_probs_dropout_prob": 0.1,
+  "attention_window": [
+    700,
+    700,
+    700,
+    700,
+    700,
+    700,
+    700,
+    700,
+    700,
+    700,
+    700,
+    700
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "O",
+    "1": "B",
+    "2": "I"
+  },
+  "ignore_attention_mask": false,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B": 1,
+    "I": 2,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 4098,
+  "model_type": "longformer",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "onnx_export": false,
+  "pad_token_id": 1,
+  "sep_token_id": 2,
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "type_vocab_size": 1,
+  "vocab_size": 50265
+}

checkpoint-648/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:863e737ed08efedcf64f132e6c5b638fc3463baa3ed60c2777b63fe2d95f7b54
+size 592318676

checkpoint-648/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e898d7e72a6f439c10848762f5f936e49071ad2297e657877f066935ed9c824
+size 1014645498

checkpoint-648/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e241cb4a4c96fc25c055fcd71173708b81cd84562a71b4d7503d9ffa5397d596
+size 14244

checkpoint-648/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17d9e668ea6c9b485e63f3e5ef87c1cfb58934d5fcbfc76edc86a350fba51ded
+size 1064

checkpoint-648/special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

checkpoint-648/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 4096,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "LongformerTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

checkpoint-648/trainer_state.json

@@ -0,0 +1,340 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 648,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_B": {
+        "f1-score": 0.8087666523420713,
+        "precision": 0.8189730200174065,
+        "recall": 0.7988115449915111,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.9374597034171502,
+        "precision": 0.9521418826739427,
+        "recall": 0.9232234509762421,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.8881362400419788,
+        "precision": 0.863303347862376,
+        "recall": 0.9144400785854617,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9154245298608586,
+      "eval_loss": 0.2228638380765915,
+      "eval_macro avg": {
+        "f1-score": 0.8781208652670669,
+        "precision": 0.8781394168512416,
+        "recall": 0.8788250248510717,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7167,
+      "eval_samples_per_second": 17.173,
+      "eval_steps_per_second": 2.332,
+      "eval_weighted avg": {
+        "f1-score": 0.9158543469268937,
+        "precision": 0.9170673807540184,
+        "recall": 0.9154245298608586,
+        "support": 30257.0
+      },
+      "step": 81
+    },
+    {
+      "epoch": 2.0,
+      "eval_B": {
+        "f1-score": 0.8401515151515152,
+        "precision": 0.7585499316005472,
+        "recall": 0.9414261460101867,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.9568235603470944,
+        "precision": 0.9510219016256338,
+        "recall": 0.9626964389650247,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9185648054827655,
+        "precision": 0.9430877483443708,
+        "recall": 0.8952848722986247,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9391876260038999,
+      "eval_loss": 0.16692574322223663,
+      "eval_macro avg": {
+        "f1-score": 0.905179960327125,
+        "precision": 0.8842198605235172,
+        "recall": 0.9331358190912787,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7116,
+      "eval_samples_per_second": 17.192,
+      "eval_steps_per_second": 2.335,
+      "eval_weighted avg": {
+        "f1-score": 0.9394089523635119,
+        "precision": 0.9408589092241132,
+        "recall": 0.9391876260038999,
+        "support": 30257.0
+      },
+      "step": 162
+    },
+    {
+      "epoch": 3.0,
+      "eval_B": {
+        "f1-score": 0.8612244897959184,
+        "precision": 0.8294025157232704,
+        "recall": 0.8955857385398981,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.951106371634231,
+        "precision": 0.9584654236741712,
+        "recall": 0.9438594634636753,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9106743212999903,
+        "precision": 0.9021592442645074,
+        "recall": 0.9193516699410609,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9337343424662061,
+      "eval_loss": 0.1742253601551056,
+      "eval_macro avg": {
+        "f1-score": 0.9076683942433799,
+        "precision": 0.8966757278873163,
+        "recall": 0.9195989573148781,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7094,
+      "eval_samples_per_second": 17.2,
+      "eval_steps_per_second": 2.336,
+      "eval_weighted avg": {
+        "f1-score": 0.9340035812317422,
+        "precision": 0.9344963252190521,
+        "recall": 0.9337343424662061,
+        "support": 30257.0
+      },
+      "step": 243
+    },
+    {
+      "epoch": 4.0,
+      "eval_B": {
+        "f1-score": 0.8717948717948719,
+        "precision": 0.8254931714719271,
+        "recall": 0.9235993208828522,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.9561271586371415,
+        "precision": 0.9374586871408959,
+        "recall": 0.9755542621302714,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9117828500925355,
+        "precision": 0.956427955133736,
+        "recall": 0.8711198428290766,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9383944211256899,
+      "eval_loss": 0.18792153894901276,
+      "eval_macro avg": {
+        "f1-score": 0.9132349601748496,
+        "precision": 0.9064599379155197,
+        "recall": 0.9234244752807333,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7342,
+      "eval_samples_per_second": 17.11,
+      "eval_steps_per_second": 2.324,
+      "eval_weighted avg": {
+        "f1-score": 0.9379241479327001,
+        "precision": 0.9394817486046585,
+        "recall": 0.9383944211256899,
+        "support": 30257.0
+      },
+      "step": 324
+    },
+    {
+      "epoch": 5.0,
+      "eval_B": {
+        "f1-score": 0.8742949234488315,
+        "precision": 0.8320552147239264,
+        "recall": 0.9210526315789473,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.9583957528450158,
+        "precision": 0.9521148825065274,
+        "recall": 0.9647600402137679,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9198818996146724,
+        "precision": 0.9375701315923697,
+        "recall": 0.9028487229862475,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9422282447037049,
+      "eval_loss": 0.19990424811840057,
+      "eval_macro avg": {
+        "f1-score": 0.9175241919695066,
+        "precision": 0.9072467429409411,
+        "recall": 0.9295537982596542,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7101,
+      "eval_samples_per_second": 17.197,
+      "eval_steps_per_second": 2.335,
+      "eval_weighted avg": {
+        "f1-score": 0.9421634164298524,
+        "precision": 0.9425469857238316,
+        "recall": 0.9422282447037049,
+        "support": 30257.0
+      },
+      "step": 405
+    },
+    {
+      "epoch": 6.0,
+      "eval_B": {
+        "f1-score": 0.8808247422680412,
+        "precision": 0.8564554931836408,
+        "recall": 0.9066213921901528,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.957406827435014,
+        "precision": 0.9446361435855178,
+        "recall": 0.9705275411397428,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9163000050574014,
+        "precision": 0.9443344105076619,
+        "recall": 0.8898821218074656,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9409062365733549,
+      "eval_loss": 0.20615626871585846,
+      "eval_macro avg": {
+        "f1-score": 0.9181771915868189,
+        "precision": 0.9151420157589403,
+        "recall": 0.922343685045787,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.6949,
+      "eval_samples_per_second": 17.253,
+      "eval_steps_per_second": 2.343,
+      "eval_weighted avg": {
+        "f1-score": 0.9405948120953971,
+        "precision": 0.941101475610967,
+        "recall": 0.9409062365733549,
+        "support": 30257.0
+      },
+      "step": 486
+    },
+    {
+      "epoch": 6.17,
+      "grad_norm": 2.4104158878326416,
+      "learning_rate": 1.7530864197530865e-05,
+      "loss": 0.1662,
+      "step": 500
+    },
+    {
+      "epoch": 7.0,
+      "eval_B": {
+        "f1-score": 0.8836820083682009,
+        "precision": 0.8712871287128713,
+        "recall": 0.8964346349745331,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.9614174269469787,
+        "precision": 0.9609854091774159,
+        "recall": 0.9618498333245146,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9271751440248166,
+        "precision": 0.9295093296475466,
+        "recall": 0.924852652259332,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9468552731599299,
+      "eval_loss": 0.22925782203674316,
+      "eval_macro avg": {
+        "f1-score": 0.9240915264466655,
+        "precision": 0.9205939558459445,
+        "recall": 0.9277123735194599,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7288,
+      "eval_samples_per_second": 17.129,
+      "eval_steps_per_second": 2.326,
+      "eval_weighted avg": {
+        "f1-score": 0.9468700903559945,
+        "precision": 0.9469030129054358,
+        "recall": 0.9468552731599299,
+        "support": 30257.0
+      },
+      "step": 567
+    },
+    {
+      "epoch": 8.0,
+      "eval_B": {
+        "f1-score": 0.8803630363036303,
+        "precision": 0.8563402889245586,
+        "recall": 0.9057724957555179,
+        "support": 1178.0
+      },
+      "eval_I": {
+        "f1-score": 0.9580651967507121,
+        "precision": 0.9550928116947994,
+        "recall": 0.9610561405365363,
+        "support": 18899.0
+      },
+      "eval_O": {
+        "f1-score": 0.9208882720333101,
+        "precision": 0.9294576746047628,
+        "recall": 0.912475442043222,
+        "support": 10180.0
+      },
+      "eval_accuracy": 0.9425587467362925,
+      "eval_loss": 0.22973769903182983,
+      "eval_macro avg": {
+        "f1-score": 0.9197721683625509,
+        "precision": 0.9136302584080402,
+        "recall": 0.9264346927784253,
+        "support": 30257.0
+      },
+      "eval_runtime": 4.7191,
+      "eval_samples_per_second": 17.164,
+      "eval_steps_per_second": 2.331,
+      "eval_weighted avg": {
+        "f1-score": 0.9425317916335554,
+        "precision": 0.942623096673485,
+        "recall": 0.9425587467362925,
+        "support": 30257.0
+      },
+      "step": 648
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4050,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "total_flos": 1146720882686400.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-648/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9f81281de988dbf2616f1c96e725c00f6e4415b289b670a7cb21136eb66e95f
+size 4856