Upload folder using huggingface_hub

.gitattributes

@@ -53,3 +53,13 @@ LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_ledgar/checkpoint-3700/tokenizer.j
 LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_ledgar/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_scotus/checkpoint-450/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_scotus/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_amazon/checkpoint-350/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_amazon/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_ledgar/checkpoint-1800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_ledgar/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_patent/checkpoint-650/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_patent/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_scotus/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_scotus/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_twitter/checkpoint-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+max_seq_length_512_experiments/google/gemma_2b_twitter/tokenizer.json filter=lfs diff=lfs merge=lfs -text

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/README.md

@@ -0,0 +1,76 @@
+---
+license: mit
+base_model: FacebookAI/roberta-base
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: roberta_base_amazon
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# roberta_base_amazon
+
+This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6534
+- Accuracy: 0.8169
+- F1 Macro: 0.7580
+- F1 Micro: 0.8169
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|
+| 2.1516        | 0.26  | 50   | 1.8448          | 0.6186   | 0.4677   | 0.6186   |
+| 1.2738        | 0.53  | 100  | 1.1242          | 0.7266   | 0.6117   | 0.7266   |
+| 0.97          | 0.79  | 150  | 0.9489          | 0.7451   | 0.6479   | 0.7451   |
+| 0.8609        | 1.05  | 200  | 0.8303          | 0.7872   | 0.7054   | 0.7872   |
+| 0.6901        | 1.32  | 250  | 0.7654          | 0.7899   | 0.7240   | 0.7899   |
+| 0.7831        | 1.58  | 300  | 0.7475          | 0.7958   | 0.7314   | 0.7958   |
+| 0.6938        | 1.84  | 350  | 0.6926          | 0.8116   | 0.7498   | 0.8116   |
+| 0.6257        | 2.11  | 400  | 0.6786          | 0.8149   | 0.7528   | 0.8149   |
+| 0.6242        | 2.37  | 450  | 0.6667          | 0.8129   | 0.7544   | 0.8129   |
+| 0.5308        | 2.63  | 500  | 0.6557          | 0.8208   | 0.7612   | 0.8208   |
+| 0.5567        | 2.89  | 550  | 0.6534          | 0.8169   | 0.7580   | 0.8169   |
+
+
+### Framework versions
+
+- Transformers 4.39.0.dev0
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/all_results.json

@@ -0,0 +1,23 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8168642951251647,
+    "eval_f1_macro": 0.7580358592296027,
+    "eval_f1_micro": 0.8168642951251647,
+    "eval_loss": 0.6533666849136353,
+    "eval_runtime": 6.1641,
+    "eval_samples": 1518,
+    "eval_samples_per_second": 246.265,
+    "eval_steps_per_second": 3.894,
+    "test_accuracy": 0.8445322793148881,
+    "test_f1_macro": 0.7842149767614086,
+    "test_f1_micro": 0.8445322793148881,
+    "test_loss": 0.6160555481910706,
+    "test_runtime": 6.1735,
+    "test_samples_per_second": 245.891,
+    "test_steps_per_second": 3.888,
+    "train_loss": 0.9715813352350603,
+    "train_runtime": 517.8127,
+    "train_samples": 12144,
+    "train_samples_per_second": 70.357,
+    "train_steps_per_second": 1.101
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/config.json

@@ -0,0 +1,79 @@
+{
+  "_name_or_path": "FacebookAI/roberta-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "text-classification",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "0",
+    "1": "1",
+    "2": "10",
+    "3": "11",
+    "4": "12",
+    "5": "13",
+    "6": "14",
+    "7": "15",
+    "8": "16",
+    "9": "17",
+    "10": "18",
+    "11": "19",
+    "12": "2",
+    "13": "20",
+    "14": "21",
+    "15": "22",
+    "16": "3",
+    "17": "4",
+    "18": "5",
+    "19": "6",
+    "20": "7",
+    "21": "8",
+    "22": "9"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "0": 0,
+    "1": 1,
+    "10": 2,
+    "11": 3,
+    "12": 4,
+    "13": 5,
+    "14": 6,
+    "15": 7,
+    "16": 8,
+    "17": 9,
+    "18": 10,
+    "19": 11,
+    "2": 12,
+    "20": 13,
+    "21": 14,
+    "22": 15,
+    "3": 16,
+    "4": 17,
+    "5": 18,
+    "6": 19,
+    "7": 20,
+    "8": 21,
+    "9": 22
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5529776b199b93ec0c80ecc9d132ba420cad62bd2776c98051955e3811806217
+size 498677420

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04d8f20a3046f3e31146ed3b10d3cc11c5f945a59b339f88add1e6beaf9a46fa
+size 997474682

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fce13561fdf9bd6b7df41528eea6b2cf2322f7ee4d0ff11a2e284f821c9f8ca
+size 14512

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18b7419cadb217caad2884bb92f5553e630b7a8bd80c1a1af566f4e372953107
+size 14512

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6304996f9341a5bf8bd2ea4d7e28f3f6c868b8f16d95b47d99cce2e178763052
+size 1064

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/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>"
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": false,
+  "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": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/trainer_state.json

@@ -0,0 +1,527 @@
+{
+  "best_metric": 0.6533666849136353,
+  "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/FacebookAI/roberta_base_amazon/checkpoint-550",
+  "epoch": 2.8947368421052633,
+  "eval_steps": 50,
+  "global_step": 550,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "grad_norm": 2.1349098682403564,
+      "learning_rate": 1.9649122807017544e-05,
+      "loss": 3.1043,
+      "step": 10
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 2.9056429862976074,
+      "learning_rate": 1.929824561403509e-05,
+      "loss": 3.0736,
+      "step": 20
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 3.7570571899414062,
+      "learning_rate": 1.894736842105263e-05,
+      "loss": 2.8836,
+      "step": 30
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 4.855740547180176,
+      "learning_rate": 1.8596491228070176e-05,
+      "loss": 2.521,
+      "step": 40
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 5.7522454261779785,
+      "learning_rate": 1.824561403508772e-05,
+      "loss": 2.1516,
+      "step": 50
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.6185770750988142,
+      "eval_f1_macro": 0.467696801300761,
+      "eval_f1_micro": 0.6185770750988142,
+      "eval_loss": 1.844839334487915,
+      "eval_runtime": 6.2653,
+      "eval_samples_per_second": 242.287,
+      "eval_steps_per_second": 3.831,
+      "step": 50
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 6.644294261932373,
+      "learning_rate": 1.7894736842105264e-05,
+      "loss": 1.8889,
+      "step": 60
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 7.293979644775391,
+      "learning_rate": 1.754385964912281e-05,
+      "loss": 1.6008,
+      "step": 70
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 8.406970024108887,
+      "learning_rate": 1.719298245614035e-05,
+      "loss": 1.4477,
+      "step": 80
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 6.161045074462891,
+      "learning_rate": 1.6842105263157896e-05,
+      "loss": 1.3757,
+      "step": 90
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 8.83963680267334,
+      "learning_rate": 1.649122807017544e-05,
+      "loss": 1.2738,
+      "step": 100
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.7266139657444005,
+      "eval_f1_macro": 0.6117051683931982,
+      "eval_f1_micro": 0.7266139657444005,
+      "eval_loss": 1.1241694688796997,
+      "eval_runtime": 6.2683,
+      "eval_samples_per_second": 242.169,
+      "eval_steps_per_second": 3.829,
+      "step": 100
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 6.5718607902526855,
+      "learning_rate": 1.6140350877192984e-05,
+      "loss": 1.1285,
+      "step": 110
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 7.004693031311035,
+      "learning_rate": 1.578947368421053e-05,
+      "loss": 1.101,
+      "step": 120
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 9.352388381958008,
+      "learning_rate": 1.543859649122807e-05,
+      "loss": 1.0951,
+      "step": 130
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 9.520184516906738,
+      "learning_rate": 1.5087719298245615e-05,
+      "loss": 1.0299,
+      "step": 140
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 8.307791709899902,
+      "learning_rate": 1.4736842105263159e-05,
+      "loss": 0.97,
+      "step": 150
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.7450592885375494,
+      "eval_f1_macro": 0.6478814545129535,
+      "eval_f1_micro": 0.7450592885375494,
+      "eval_loss": 0.9488905072212219,
+      "eval_runtime": 6.267,
+      "eval_samples_per_second": 242.22,
+      "eval_steps_per_second": 3.83,
+      "step": 150
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 8.785259246826172,
+      "learning_rate": 1.4385964912280704e-05,
+      "loss": 0.9838,
+      "step": 160
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 9.214316368103027,
+      "learning_rate": 1.4035087719298246e-05,
+      "loss": 0.9468,
+      "step": 170
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 10.883708000183105,
+      "learning_rate": 1.3684210526315791e-05,
+      "loss": 0.9317,
+      "step": 180
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 9.814682960510254,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.949,
+      "step": 190
+    },
+    {
+      "epoch": 1.05,
+      "grad_norm": 7.213553428649902,
+      "learning_rate": 1.2982456140350879e-05,
+      "loss": 0.8609,
+      "step": 200
+    },
+    {
+      "epoch": 1.05,
+      "eval_accuracy": 0.7872200263504612,
+      "eval_f1_macro": 0.7054358382105227,
+      "eval_f1_micro": 0.7872200263504612,
+      "eval_loss": 0.8302952647209167,
+      "eval_runtime": 6.2442,
+      "eval_samples_per_second": 243.106,
+      "eval_steps_per_second": 3.844,
+      "step": 200
+    },
+    {
+      "epoch": 1.11,
+      "grad_norm": 8.742850303649902,
+      "learning_rate": 1.263157894736842e-05,
+      "loss": 0.7843,
+      "step": 210
+    },
+    {
+      "epoch": 1.16,
+      "grad_norm": 13.062145233154297,
+      "learning_rate": 1.2280701754385966e-05,
+      "loss": 0.8302,
+      "step": 220
+    },
+    {
+      "epoch": 1.21,
+      "grad_norm": 8.80545425415039,
+      "learning_rate": 1.192982456140351e-05,
+      "loss": 0.8087,
+      "step": 230
+    },
+    {
+      "epoch": 1.26,
+      "grad_norm": 9.45230770111084,
+      "learning_rate": 1.1578947368421053e-05,
+      "loss": 0.7372,
+      "step": 240
+    },
+    {
+      "epoch": 1.32,
+      "grad_norm": 8.2438383102417,
+      "learning_rate": 1.1228070175438597e-05,
+      "loss": 0.6901,
+      "step": 250
+    },
+    {
+      "epoch": 1.32,
+      "eval_accuracy": 0.7898550724637681,
+      "eval_f1_macro": 0.7240233789782649,
+      "eval_f1_micro": 0.7898550724637681,
+      "eval_loss": 0.765446662902832,
+      "eval_runtime": 6.2385,
+      "eval_samples_per_second": 243.326,
+      "eval_steps_per_second": 3.847,
+      "step": 250
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 9.07975959777832,
+      "learning_rate": 1.0877192982456142e-05,
+      "loss": 0.7008,
+      "step": 260
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 7.3508148193359375,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.7267,
+      "step": 270
+    },
+    {
+      "epoch": 1.47,
+      "grad_norm": 13.495626449584961,
+      "learning_rate": 1.017543859649123e-05,
+      "loss": 0.7384,
+      "step": 280
+    },
+    {
+      "epoch": 1.53,
+      "grad_norm": 9.769052505493164,
+      "learning_rate": 9.824561403508772e-06,
+      "loss": 0.6653,
+      "step": 290
+    },
+    {
+      "epoch": 1.58,
+      "grad_norm": 9.317341804504395,
+      "learning_rate": 9.473684210526315e-06,
+      "loss": 0.7831,
+      "step": 300
+    },
+    {
+      "epoch": 1.58,
+      "eval_accuracy": 0.7957839262187089,
+      "eval_f1_macro": 0.7313867200978218,
+      "eval_f1_micro": 0.7957839262187089,
+      "eval_loss": 0.7475197315216064,
+      "eval_runtime": 6.2231,
+      "eval_samples_per_second": 243.929,
+      "eval_steps_per_second": 3.857,
+      "step": 300
+    },
+    {
+      "epoch": 1.63,
+      "grad_norm": 9.545998573303223,
+      "learning_rate": 9.12280701754386e-06,
+      "loss": 0.7343,
+      "step": 310
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 9.613531112670898,
+      "learning_rate": 8.771929824561405e-06,
+      "loss": 0.6921,
+      "step": 320
+    },
+    {
+      "epoch": 1.74,
+      "grad_norm": 8.262693405151367,
+      "learning_rate": 8.421052631578948e-06,
+      "loss": 0.7007,
+      "step": 330
+    },
+    {
+      "epoch": 1.79,
+      "grad_norm": 13.85246467590332,
+      "learning_rate": 8.070175438596492e-06,
+      "loss": 0.7754,
+      "step": 340
+    },
+    {
+      "epoch": 1.84,
+      "grad_norm": 11.566596031188965,
+      "learning_rate": 7.719298245614036e-06,
+      "loss": 0.6938,
+      "step": 350
+    },
+    {
+      "epoch": 1.84,
+      "eval_accuracy": 0.8115942028985508,
+      "eval_f1_macro": 0.7497539278629068,
+      "eval_f1_micro": 0.8115942028985508,
+      "eval_loss": 0.6926287412643433,
+      "eval_runtime": 6.2112,
+      "eval_samples_per_second": 244.396,
+      "eval_steps_per_second": 3.864,
+      "step": 350
+    },
+    {
+      "epoch": 1.89,
+      "grad_norm": 12.794224739074707,
+      "learning_rate": 7.368421052631579e-06,
+      "loss": 0.6381,
+      "step": 360
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 11.993423461914062,
+      "learning_rate": 7.017543859649123e-06,
+      "loss": 0.6974,
+      "step": 370
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 13.328503608703613,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.6957,
+      "step": 380
+    },
+    {
+      "epoch": 2.05,
+      "grad_norm": 7.529949188232422,
+      "learning_rate": 6.31578947368421e-06,
+      "loss": 0.6489,
+      "step": 390
+    },
+    {
+      "epoch": 2.11,
+      "grad_norm": 7.078703880310059,
+      "learning_rate": 5.964912280701755e-06,
+      "loss": 0.6257,
+      "step": 400
+    },
+    {
+      "epoch": 2.11,
+      "eval_accuracy": 0.8148880105401844,
+      "eval_f1_macro": 0.7527719650823338,
+      "eval_f1_micro": 0.8148880105401844,
+      "eval_loss": 0.6785778403282166,
+      "eval_runtime": 6.2049,
+      "eval_samples_per_second": 244.644,
+      "eval_steps_per_second": 3.868,
+      "step": 400
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 9.676349639892578,
+      "learning_rate": 5.6140350877192985e-06,
+      "loss": 0.5719,
+      "step": 410
+    },
+    {
+      "epoch": 2.21,
+      "grad_norm": 12.939827919006348,
+      "learning_rate": 5.263157894736842e-06,
+      "loss": 0.6037,
+      "step": 420
+    },
+    {
+      "epoch": 2.26,
+      "grad_norm": 11.817204475402832,
+      "learning_rate": 4.912280701754386e-06,
+      "loss": 0.5931,
+      "step": 430
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 9.458736419677734,
+      "learning_rate": 4.56140350877193e-06,
+      "loss": 0.6051,
+      "step": 440
+    },
+    {
+      "epoch": 2.37,
+      "grad_norm": 6.873404502868652,
+      "learning_rate": 4.210526315789474e-06,
+      "loss": 0.6242,
+      "step": 450
+    },
+    {
+      "epoch": 2.37,
+      "eval_accuracy": 0.8129117259552042,
+      "eval_f1_macro": 0.7543580436769726,
+      "eval_f1_micro": 0.8129117259552042,
+      "eval_loss": 0.6666637063026428,
+      "eval_runtime": 6.2079,
+      "eval_samples_per_second": 244.529,
+      "eval_steps_per_second": 3.866,
+      "step": 450
+    },
+    {
+      "epoch": 2.42,
+      "grad_norm": 12.340581893920898,
+      "learning_rate": 3.859649122807018e-06,
+      "loss": 0.5459,
+      "step": 460
+    },
+    {
+      "epoch": 2.47,
+      "grad_norm": 10.073423385620117,
+      "learning_rate": 3.5087719298245615e-06,
+      "loss": 0.5739,
+      "step": 470
+    },
+    {
+      "epoch": 2.53,
+      "grad_norm": 13.684353828430176,
+      "learning_rate": 3.157894736842105e-06,
+      "loss": 0.6161,
+      "step": 480
+    },
+    {
+      "epoch": 2.58,
+      "grad_norm": 12.562374114990234,
+      "learning_rate": 2.8070175438596493e-06,
+      "loss": 0.5231,
+      "step": 490
+    },
+    {
+      "epoch": 2.63,
+      "grad_norm": 10.13200569152832,
+      "learning_rate": 2.456140350877193e-06,
+      "loss": 0.5308,
+      "step": 500
+    },
+    {
+      "epoch": 2.63,
+      "eval_accuracy": 0.8208168642951251,
+      "eval_f1_macro": 0.7612436541002552,
+      "eval_f1_micro": 0.8208168642951251,
+      "eval_loss": 0.6556726098060608,
+      "eval_runtime": 6.2,
+      "eval_samples_per_second": 244.84,
+      "eval_steps_per_second": 3.871,
+      "step": 500
+    },
+    {
+      "epoch": 2.68,
+      "grad_norm": 11.53864860534668,
+      "learning_rate": 2.105263157894737e-06,
+      "loss": 0.533,
+      "step": 510
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 14.059792518615723,
+      "learning_rate": 1.7543859649122807e-06,
+      "loss": 0.5548,
+      "step": 520
+    },
+    {
+      "epoch": 2.79,
+      "grad_norm": 11.259710311889648,
+      "learning_rate": 1.4035087719298246e-06,
+      "loss": 0.539,
+      "step": 530
+    },
+    {
+      "epoch": 2.84,
+      "grad_norm": 8.336936950683594,
+      "learning_rate": 1.0526315789473685e-06,
+      "loss": 0.5409,
+      "step": 540
+    },
+    {
+      "epoch": 2.89,
+      "grad_norm": 7.9991655349731445,
+      "learning_rate": 7.017543859649123e-07,
+      "loss": 0.5567,
+      "step": 550
+    },
+    {
+      "epoch": 2.89,
+      "eval_accuracy": 0.8168642951251647,
+      "eval_f1_macro": 0.7580358592296027,
+      "eval_f1_micro": 0.8168642951251647,
+      "eval_loss": 0.6533666849136353,
+      "eval_runtime": 6.2016,
+      "eval_samples_per_second": 244.777,
+      "eval_steps_per_second": 3.87,
+      "step": 550
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 570,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "total_flos": 9263255692247040.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/checkpoint-550/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e50be805655392885abb404908f71df5c35cf43658730dbb84159c8ba0987945
+size 5048

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/config.json

@@ -0,0 +1,79 @@
+{
+  "_name_or_path": "FacebookAI/roberta-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "text-classification",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "0",
+    "1": "1",
+    "2": "10",
+    "3": "11",
+    "4": "12",
+    "5": "13",
+    "6": "14",
+    "7": "15",
+    "8": "16",
+    "9": "17",
+    "10": "18",
+    "11": "19",
+    "12": "2",
+    "13": "20",
+    "14": "21",
+    "15": "22",
+    "16": "3",
+    "17": "4",
+    "18": "5",
+    "19": "6",
+    "20": "7",
+    "21": "8",
+    "22": "9"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "0": 0,
+    "1": 1,
+    "10": 2,
+    "11": 3,
+    "12": 4,
+    "13": 5,
+    "14": 6,
+    "15": 7,
+    "16": 8,
+    "17": 9,
+    "18": 10,
+    "19": 11,
+    "2": 12,
+    "20": 13,
+    "21": 14,
+    "22": 15,
+    "3": 16,
+    "4": 17,
+    "5": 18,
+    "6": 19,
+    "7": 20,
+    "8": 21,
+    "9": 22
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8168642951251647,
+    "eval_f1_macro": 0.7580358592296027,
+    "eval_f1_micro": 0.8168642951251647,
+    "eval_loss": 0.6533666849136353,
+    "eval_runtime": 6.1641,
+    "eval_samples": 1518,
+    "eval_samples_per_second": 246.265,
+    "eval_steps_per_second": 3.894
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5529776b199b93ec0c80ecc9d132ba420cad62bd2776c98051955e3811806217
+size 498677420

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/run.log

@@ -0,0 +1,4 @@
+03/20/2024 18:52:55 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, 16-bits training: False
+03/20/2024 18:52:55 - WARNING - __main__ - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, 16-bits training: False
+03/20/2024 18:52:59 - WARNING - __main__ - The label2id key in the model config.json is not equal to the label2id key of this run. You can ignore this if you are doing finetuning.
+03/20/2024 18:52:59 - WARNING - __main__ - The label2id key in the model config.json is not equal to the label2id key of this run. You can ignore this if you are doing finetuning.

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/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>"
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/test_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "test_accuracy": 0.8445322793148881,
+    "test_f1_macro": 0.7842149767614086,
+    "test_f1_micro": 0.8445322793148881,
+    "test_loss": 0.6160555481910706,
+    "test_runtime": 6.1735,
+    "test_samples_per_second": 245.891,
+    "test_steps_per_second": 3.888
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": false,
+  "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": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.9715813352350603,
+    "train_runtime": 517.8127,
+    "train_samples": 12144,
+    "train_samples_per_second": 70.357,
+    "train_steps_per_second": 1.101
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/trainer_state.json

@@ -0,0 +1,550 @@
+{
+  "best_metric": 0.6533666849136353,
+  "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/FacebookAI/roberta_base_amazon/checkpoint-550",
+  "epoch": 3.0,
+  "eval_steps": 50,
+  "global_step": 570,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "grad_norm": 2.1349098682403564,
+      "learning_rate": 1.9649122807017544e-05,
+      "loss": 3.1043,
+      "step": 10
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 2.9056429862976074,
+      "learning_rate": 1.929824561403509e-05,
+      "loss": 3.0736,
+      "step": 20
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 3.7570571899414062,
+      "learning_rate": 1.894736842105263e-05,
+      "loss": 2.8836,
+      "step": 30
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 4.855740547180176,
+      "learning_rate": 1.8596491228070176e-05,
+      "loss": 2.521,
+      "step": 40
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 5.7522454261779785,
+      "learning_rate": 1.824561403508772e-05,
+      "loss": 2.1516,
+      "step": 50
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.6185770750988142,
+      "eval_f1_macro": 0.467696801300761,
+      "eval_f1_micro": 0.6185770750988142,
+      "eval_loss": 1.844839334487915,
+      "eval_runtime": 6.2653,
+      "eval_samples_per_second": 242.287,
+      "eval_steps_per_second": 3.831,
+      "step": 50
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 6.644294261932373,
+      "learning_rate": 1.7894736842105264e-05,
+      "loss": 1.8889,
+      "step": 60
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 7.293979644775391,
+      "learning_rate": 1.754385964912281e-05,
+      "loss": 1.6008,
+      "step": 70
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 8.406970024108887,
+      "learning_rate": 1.719298245614035e-05,
+      "loss": 1.4477,
+      "step": 80
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 6.161045074462891,
+      "learning_rate": 1.6842105263157896e-05,
+      "loss": 1.3757,
+      "step": 90
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 8.83963680267334,
+      "learning_rate": 1.649122807017544e-05,
+      "loss": 1.2738,
+      "step": 100
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.7266139657444005,
+      "eval_f1_macro": 0.6117051683931982,
+      "eval_f1_micro": 0.7266139657444005,
+      "eval_loss": 1.1241694688796997,
+      "eval_runtime": 6.2683,
+      "eval_samples_per_second": 242.169,
+      "eval_steps_per_second": 3.829,
+      "step": 100
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 6.5718607902526855,
+      "learning_rate": 1.6140350877192984e-05,
+      "loss": 1.1285,
+      "step": 110
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 7.004693031311035,
+      "learning_rate": 1.578947368421053e-05,
+      "loss": 1.101,
+      "step": 120
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 9.352388381958008,
+      "learning_rate": 1.543859649122807e-05,
+      "loss": 1.0951,
+      "step": 130
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 9.520184516906738,
+      "learning_rate": 1.5087719298245615e-05,
+      "loss": 1.0299,
+      "step": 140
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 8.307791709899902,
+      "learning_rate": 1.4736842105263159e-05,
+      "loss": 0.97,
+      "step": 150
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.7450592885375494,
+      "eval_f1_macro": 0.6478814545129535,
+      "eval_f1_micro": 0.7450592885375494,
+      "eval_loss": 0.9488905072212219,
+      "eval_runtime": 6.267,
+      "eval_samples_per_second": 242.22,
+      "eval_steps_per_second": 3.83,
+      "step": 150
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 8.785259246826172,
+      "learning_rate": 1.4385964912280704e-05,
+      "loss": 0.9838,
+      "step": 160
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 9.214316368103027,
+      "learning_rate": 1.4035087719298246e-05,
+      "loss": 0.9468,
+      "step": 170
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 10.883708000183105,
+      "learning_rate": 1.3684210526315791e-05,
+      "loss": 0.9317,
+      "step": 180
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 9.814682960510254,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.949,
+      "step": 190
+    },
+    {
+      "epoch": 1.05,
+      "grad_norm": 7.213553428649902,
+      "learning_rate": 1.2982456140350879e-05,
+      "loss": 0.8609,
+      "step": 200
+    },
+    {
+      "epoch": 1.05,
+      "eval_accuracy": 0.7872200263504612,
+      "eval_f1_macro": 0.7054358382105227,
+      "eval_f1_micro": 0.7872200263504612,
+      "eval_loss": 0.8302952647209167,
+      "eval_runtime": 6.2442,
+      "eval_samples_per_second": 243.106,
+      "eval_steps_per_second": 3.844,
+      "step": 200
+    },
+    {
+      "epoch": 1.11,
+      "grad_norm": 8.742850303649902,
+      "learning_rate": 1.263157894736842e-05,
+      "loss": 0.7843,
+      "step": 210
+    },
+    {
+      "epoch": 1.16,
+      "grad_norm": 13.062145233154297,
+      "learning_rate": 1.2280701754385966e-05,
+      "loss": 0.8302,
+      "step": 220
+    },
+    {
+      "epoch": 1.21,
+      "grad_norm": 8.80545425415039,
+      "learning_rate": 1.192982456140351e-05,
+      "loss": 0.8087,
+      "step": 230
+    },
+    {
+      "epoch": 1.26,
+      "grad_norm": 9.45230770111084,
+      "learning_rate": 1.1578947368421053e-05,
+      "loss": 0.7372,
+      "step": 240
+    },
+    {
+      "epoch": 1.32,
+      "grad_norm": 8.2438383102417,
+      "learning_rate": 1.1228070175438597e-05,
+      "loss": 0.6901,
+      "step": 250
+    },
+    {
+      "epoch": 1.32,
+      "eval_accuracy": 0.7898550724637681,
+      "eval_f1_macro": 0.7240233789782649,
+      "eval_f1_micro": 0.7898550724637681,
+      "eval_loss": 0.765446662902832,
+      "eval_runtime": 6.2385,
+      "eval_samples_per_second": 243.326,
+      "eval_steps_per_second": 3.847,
+      "step": 250
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 9.07975959777832,
+      "learning_rate": 1.0877192982456142e-05,
+      "loss": 0.7008,
+      "step": 260
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 7.3508148193359375,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.7267,
+      "step": 270
+    },
+    {
+      "epoch": 1.47,
+      "grad_norm": 13.495626449584961,
+      "learning_rate": 1.017543859649123e-05,
+      "loss": 0.7384,
+      "step": 280
+    },
+    {
+      "epoch": 1.53,
+      "grad_norm": 9.769052505493164,
+      "learning_rate": 9.824561403508772e-06,
+      "loss": 0.6653,
+      "step": 290
+    },
+    {
+      "epoch": 1.58,
+      "grad_norm": 9.317341804504395,
+      "learning_rate": 9.473684210526315e-06,
+      "loss": 0.7831,
+      "step": 300
+    },
+    {
+      "epoch": 1.58,
+      "eval_accuracy": 0.7957839262187089,
+      "eval_f1_macro": 0.7313867200978218,
+      "eval_f1_micro": 0.7957839262187089,
+      "eval_loss": 0.7475197315216064,
+      "eval_runtime": 6.2231,
+      "eval_samples_per_second": 243.929,
+      "eval_steps_per_second": 3.857,
+      "step": 300
+    },
+    {
+      "epoch": 1.63,
+      "grad_norm": 9.545998573303223,
+      "learning_rate": 9.12280701754386e-06,
+      "loss": 0.7343,
+      "step": 310
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 9.613531112670898,
+      "learning_rate": 8.771929824561405e-06,
+      "loss": 0.6921,
+      "step": 320
+    },
+    {
+      "epoch": 1.74,
+      "grad_norm": 8.262693405151367,
+      "learning_rate": 8.421052631578948e-06,
+      "loss": 0.7007,
+      "step": 330
+    },
+    {
+      "epoch": 1.79,
+      "grad_norm": 13.85246467590332,
+      "learning_rate": 8.070175438596492e-06,
+      "loss": 0.7754,
+      "step": 340
+    },
+    {
+      "epoch": 1.84,
+      "grad_norm": 11.566596031188965,
+      "learning_rate": 7.719298245614036e-06,
+      "loss": 0.6938,
+      "step": 350
+    },
+    {
+      "epoch": 1.84,
+      "eval_accuracy": 0.8115942028985508,
+      "eval_f1_macro": 0.7497539278629068,
+      "eval_f1_micro": 0.8115942028985508,
+      "eval_loss": 0.6926287412643433,
+      "eval_runtime": 6.2112,
+      "eval_samples_per_second": 244.396,
+      "eval_steps_per_second": 3.864,
+      "step": 350
+    },
+    {
+      "epoch": 1.89,
+      "grad_norm": 12.794224739074707,
+      "learning_rate": 7.368421052631579e-06,
+      "loss": 0.6381,
+      "step": 360
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 11.993423461914062,
+      "learning_rate": 7.017543859649123e-06,
+      "loss": 0.6974,
+      "step": 370
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 13.328503608703613,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.6957,
+      "step": 380
+    },
+    {
+      "epoch": 2.05,
+      "grad_norm": 7.529949188232422,
+      "learning_rate": 6.31578947368421e-06,
+      "loss": 0.6489,
+      "step": 390
+    },
+    {
+      "epoch": 2.11,
+      "grad_norm": 7.078703880310059,
+      "learning_rate": 5.964912280701755e-06,
+      "loss": 0.6257,
+      "step": 400
+    },
+    {
+      "epoch": 2.11,
+      "eval_accuracy": 0.8148880105401844,
+      "eval_f1_macro": 0.7527719650823338,
+      "eval_f1_micro": 0.8148880105401844,
+      "eval_loss": 0.6785778403282166,
+      "eval_runtime": 6.2049,
+      "eval_samples_per_second": 244.644,
+      "eval_steps_per_second": 3.868,
+      "step": 400
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 9.676349639892578,
+      "learning_rate": 5.6140350877192985e-06,
+      "loss": 0.5719,
+      "step": 410
+    },
+    {
+      "epoch": 2.21,
+      "grad_norm": 12.939827919006348,
+      "learning_rate": 5.263157894736842e-06,
+      "loss": 0.6037,
+      "step": 420
+    },
+    {
+      "epoch": 2.26,
+      "grad_norm": 11.817204475402832,
+      "learning_rate": 4.912280701754386e-06,
+      "loss": 0.5931,
+      "step": 430
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 9.458736419677734,
+      "learning_rate": 4.56140350877193e-06,
+      "loss": 0.6051,
+      "step": 440
+    },
+    {
+      "epoch": 2.37,
+      "grad_norm": 6.873404502868652,
+      "learning_rate": 4.210526315789474e-06,
+      "loss": 0.6242,
+      "step": 450
+    },
+    {
+      "epoch": 2.37,
+      "eval_accuracy": 0.8129117259552042,
+      "eval_f1_macro": 0.7543580436769726,
+      "eval_f1_micro": 0.8129117259552042,
+      "eval_loss": 0.6666637063026428,
+      "eval_runtime": 6.2079,
+      "eval_samples_per_second": 244.529,
+      "eval_steps_per_second": 3.866,
+      "step": 450
+    },
+    {
+      "epoch": 2.42,
+      "grad_norm": 12.340581893920898,
+      "learning_rate": 3.859649122807018e-06,
+      "loss": 0.5459,
+      "step": 460
+    },
+    {
+      "epoch": 2.47,
+      "grad_norm": 10.073423385620117,
+      "learning_rate": 3.5087719298245615e-06,
+      "loss": 0.5739,
+      "step": 470
+    },
+    {
+      "epoch": 2.53,
+      "grad_norm": 13.684353828430176,
+      "learning_rate": 3.157894736842105e-06,
+      "loss": 0.6161,
+      "step": 480
+    },
+    {
+      "epoch": 2.58,
+      "grad_norm": 12.562374114990234,
+      "learning_rate": 2.8070175438596493e-06,
+      "loss": 0.5231,
+      "step": 490
+    },
+    {
+      "epoch": 2.63,
+      "grad_norm": 10.13200569152832,
+      "learning_rate": 2.456140350877193e-06,
+      "loss": 0.5308,
+      "step": 500
+    },
+    {
+      "epoch": 2.63,
+      "eval_accuracy": 0.8208168642951251,
+      "eval_f1_macro": 0.7612436541002552,
+      "eval_f1_micro": 0.8208168642951251,
+      "eval_loss": 0.6556726098060608,
+      "eval_runtime": 6.2,
+      "eval_samples_per_second": 244.84,
+      "eval_steps_per_second": 3.871,
+      "step": 500
+    },
+    {
+      "epoch": 2.68,
+      "grad_norm": 11.53864860534668,
+      "learning_rate": 2.105263157894737e-06,
+      "loss": 0.533,
+      "step": 510
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 14.059792518615723,
+      "learning_rate": 1.7543859649122807e-06,
+      "loss": 0.5548,
+      "step": 520
+    },
+    {
+      "epoch": 2.79,
+      "grad_norm": 11.259710311889648,
+      "learning_rate": 1.4035087719298246e-06,
+      "loss": 0.539,
+      "step": 530
+    },
+    {
+      "epoch": 2.84,
+      "grad_norm": 8.336936950683594,
+      "learning_rate": 1.0526315789473685e-06,
+      "loss": 0.5409,
+      "step": 540
+    },
+    {
+      "epoch": 2.89,
+      "grad_norm": 7.9991655349731445,
+      "learning_rate": 7.017543859649123e-07,
+      "loss": 0.5567,
+      "step": 550
+    },
+    {
+      "epoch": 2.89,
+      "eval_accuracy": 0.8168642951251647,
+      "eval_f1_macro": 0.7580358592296027,
+      "eval_f1_micro": 0.8168642951251647,
+      "eval_loss": 0.6533666849136353,
+      "eval_runtime": 6.2016,
+      "eval_samples_per_second": 244.777,
+      "eval_steps_per_second": 3.87,
+      "step": 550
+    },
+    {
+      "epoch": 2.95,
+      "grad_norm": 10.723379135131836,
+      "learning_rate": 3.5087719298245616e-07,
+      "loss": 0.5776,
+      "step": 560
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 9.620089530944824,
+      "learning_rate": 0.0,
+      "loss": 0.6057,
+      "step": 570
+    },
+    {
+      "epoch": 3.0,
+      "step": 570,
+      "total_flos": 9600101353783296.0,
+      "train_loss": 0.9715813352350603,
+      "train_runtime": 517.8127,
+      "train_samples_per_second": 70.357,
+      "train_steps_per_second": 1.101
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 570,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "total_flos": 9600101353783296.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_amazon/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e50be805655392885abb404908f71df5c35cf43658730dbb84159c8ba0987945
+size 5048

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/README.md

@@ -0,0 +1,93 @@
+---
+license: mit
+base_model: FacebookAI/roberta-base
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: roberta_base_ledgar
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# roberta_base_ledgar
+
+This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5945
+- Accuracy: 0.8453
+- F1 Macro: 0.7392
+- F1 Micro: 0.8453
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|
+| 3.1362        | 0.11  | 100  | 2.7513          | 0.5507   | 0.2743   | 0.5507   |
+| 2.0094        | 0.21  | 200  | 1.8194          | 0.678    | 0.4121   | 0.678    |
+| 1.573         | 0.32  | 300  | 1.4166          | 0.7322   | 0.5072   | 0.7322   |
+| 1.2789        | 0.43  | 400  | 1.1782          | 0.7693   | 0.5711   | 0.7693   |
+| 1.1898        | 0.53  | 500  | 1.0305          | 0.7877   | 0.6067   | 0.7877   |
+| 1.0704        | 0.64  | 600  | 0.9311          | 0.7986   | 0.6314   | 0.7986   |
+| 0.9539        | 0.75  | 700  | 0.8658          | 0.8014   | 0.6332   | 0.8014   |
+| 0.9228        | 0.85  | 800  | 0.8173          | 0.8077   | 0.6463   | 0.8077   |
+| 0.8447        | 0.96  | 900  | 0.7711          | 0.8205   | 0.6760   | 0.8205   |
+| 0.7568        | 1.07  | 1000 | 0.7531          | 0.8197   | 0.6750   | 0.8197   |
+| 0.7484        | 1.17  | 1100 | 0.7310          | 0.8219   | 0.6822   | 0.8219   |
+| 0.7915        | 1.28  | 1200 | 0.7162          | 0.8271   | 0.6897   | 0.8271   |
+| 0.7083        | 1.39  | 1300 | 0.6928          | 0.8256   | 0.6919   | 0.8256   |
+| 0.6292        | 1.49  | 1400 | 0.6823          | 0.8281   | 0.6930   | 0.8281   |
+| 0.6646        | 1.6   | 1500 | 0.6709          | 0.8321   | 0.7047   | 0.8321   |
+| 0.6808        | 1.71  | 1600 | 0.6590          | 0.8348   | 0.7160   | 0.8348   |
+| 0.6748        | 1.81  | 1700 | 0.6456          | 0.8336   | 0.7130   | 0.8336   |
+| 0.6251        | 1.92  | 1800 | 0.6295          | 0.8377   | 0.7167   | 0.8377   |
+| 0.6155        | 2.03  | 1900 | 0.6280          | 0.8382   | 0.7147   | 0.8382   |
+| 0.5387        | 2.13  | 2000 | 0.6257          | 0.8387   | 0.7213   | 0.8387   |
+| 0.5859        | 2.24  | 2100 | 0.6193          | 0.8402   | 0.7276   | 0.8402   |
+| 0.5663        | 2.35  | 2200 | 0.6136          | 0.843    | 0.7314   | 0.843    |
+| 0.5836        | 2.45  | 2300 | 0.6110          | 0.8433   | 0.7320   | 0.8433   |
+| 0.5451        | 2.56  | 2400 | 0.5996          | 0.8463   | 0.7385   | 0.8463   |
+| 0.5198        | 2.67  | 2500 | 0.5996          | 0.8445   | 0.7385   | 0.8445   |
+| 0.5508        | 2.77  | 2600 | 0.5973          | 0.8439   | 0.7371   | 0.8439   |
+| 0.5199        | 2.88  | 2700 | 0.5945          | 0.8453   | 0.7392   | 0.8453   |
+| 0.5495        | 2.99  | 2800 | 0.5946          | 0.8451   | 0.7383   | 0.8451   |
+
+
+### Framework versions
+
+- Transformers 4.39.0.dev0
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/all_results.json

@@ -0,0 +1,23 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8453,
+    "eval_f1_macro": 0.7391787821862769,
+    "eval_f1_micro": 0.8453,
+    "eval_loss": 0.5945026874542236,
+    "eval_runtime": 40.1416,
+    "eval_samples": 10000,
+    "eval_samples_per_second": 249.118,
+    "eval_steps_per_second": 3.911,
+    "test_accuracy": 0.8466,
+    "test_f1_macro": 0.7313572079408452,
+    "test_f1_micro": 0.8466,
+    "test_loss": 0.608473002910614,
+    "test_runtime": 40.5608,
+    "test_samples_per_second": 246.543,
+    "test_steps_per_second": 3.871,
+    "train_loss": 0.9456861351259905,
+    "train_runtime": 3281.5845,
+    "train_samples": 60000,
+    "train_samples_per_second": 54.852,
+    "train_steps_per_second": 0.858
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/config.json

@@ -0,0 +1,233 @@
+{
+  "_name_or_path": "FacebookAI/roberta-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "text-classification",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "0",
+    "1": "1",
+    "2": "10",
+    "3": "11",
+    "4": "12",
+    "5": "13",
+    "6": "14",
+    "7": "15",
+    "8": "16",
+    "9": "17",
+    "10": "18",
+    "11": "19",
+    "12": "2",
+    "13": "20",
+    "14": "21",
+    "15": "22",
+    "16": "23",
+    "17": "24",
+    "18": "25",
+    "19": "26",
+    "20": "27",
+    "21": "28",
+    "22": "29",
+    "23": "3",
+    "24": "30",
+    "25": "31",
+    "26": "32",
+    "27": "33",
+    "28": "34",
+    "29": "35",
+    "30": "36",
+    "31": "37",
+    "32": "38",
+    "33": "39",
+    "34": "4",
+    "35": "40",
+    "36": "41",
+    "37": "42",
+    "38": "43",
+    "39": "44",
+    "40": "45",
+    "41": "46",
+    "42": "47",
+    "43": "48",
+    "44": "49",
+    "45": "5",
+    "46": "50",
+    "47": "51",
+    "48": "52",
+    "49": "53",
+    "50": "54",
+    "51": "55",
+    "52": "56",
+    "53": "57",
+    "54": "58",
+    "55": "59",
+    "56": "6",
+    "57": "60",
+    "58": "61",
+    "59": "62",
+    "60": "63",
+    "61": "64",
+    "62": "65",
+    "63": "66",
+    "64": "67",
+    "65": "68",
+    "66": "69",
+    "67": "7",
+    "68": "70",
+    "69": "71",
+    "70": "72",
+    "71": "73",
+    "72": "74",
+    "73": "75",
+    "74": "76",
+    "75": "77",
+    "76": "78",
+    "77": "79",
+    "78": "8",
+    "79": "80",
+    "80": "81",
+    "81": "82",
+    "82": "83",
+    "83": "84",
+    "84": "85",
+    "85": "86",
+    "86": "87",
+    "87": "88",
+    "88": "89",
+    "89": "9",
+    "90": "90",
+    "91": "91",
+    "92": "92",
+    "93": "93",
+    "94": "94",
+    "95": "95",
+    "96": "96",
+    "97": "97",
+    "98": "98",
+    "99": "99"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "0": 0,
+    "1": 1,
+    "10": 2,
+    "11": 3,
+    "12": 4,
+    "13": 5,
+    "14": 6,
+    "15": 7,
+    "16": 8,
+    "17": 9,
+    "18": 10,
+    "19": 11,
+    "2": 12,
+    "20": 13,
+    "21": 14,
+    "22": 15,
+    "23": 16,
+    "24": 17,
+    "25": 18,
+    "26": 19,
+    "27": 20,
+    "28": 21,
+    "29": 22,
+    "3": 23,
+    "30": 24,
+    "31": 25,
+    "32": 26,
+    "33": 27,
+    "34": 28,
+    "35": 29,
+    "36": 30,
+    "37": 31,
+    "38": 32,
+    "39": 33,
+    "4": 34,
+    "40": 35,
+    "41": 36,
+    "42": 37,
+    "43": 38,
+    "44": 39,
+    "45": 40,
+    "46": 41,
+    "47": 42,
+    "48": 43,
+    "49": 44,
+    "5": 45,
+    "50": 46,
+    "51": 47,
+    "52": 48,
+    "53": 49,
+    "54": 50,
+    "55": 51,
+    "56": 52,
+    "57": 53,
+    "58": 54,
+    "59": 55,
+    "6": 56,
+    "60": 57,
+    "61": 58,
+    "62": 59,
+    "63": 60,
+    "64": 61,
+    "65": 62,
+    "66": 63,
+    "67": 64,
+    "68": 65,
+    "69": 66,
+    "7": 67,
+    "70": 68,
+    "71": 69,
+    "72": 70,
+    "73": 71,
+    "74": 72,
+    "75": 73,
+    "76": 74,
+    "77": 75,
+    "78": 76,
+    "79": 77,
+    "8": 78,
+    "80": 79,
+    "81": 80,
+    "82": 81,
+    "83": 82,
+    "84": 83,
+    "85": 84,
+    "86": 85,
+    "87": 86,
+    "88": 87,
+    "89": 88,
+    "9": 89,
+    "90": 90,
+    "91": 91,
+    "92": 92,
+    "93": 93,
+    "94": 94,
+    "95": 95,
+    "96": 96,
+    "97": 97,
+    "98": 98,
+    "99": 99
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24009191e7c7cf88a3be5f0d6c77255d3e41fe6b0bd9186ec387ce66d42c73ad
+size 498914272

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e06138b32ebca096c270fd0881a15bcf8e29ac7b6b12a57aa4c9bd74496d65c6
+size 997948410

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:baac505491123af25f66ad9cd5379f7dd77166b636ef9dd76486f76c09f24b60
+size 14512

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ed10e0f0de9897236cfe46695f9cebbcf53f19665072a2a2557672590869650
+size 14512

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e8bd9cc2dc383b9b752c5c6c75783cb0d3b410b677ab68485e206a9308f9a04
+size 1064

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/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>"
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": false,
+  "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": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/trainer_state.json

@@ -0,0 +1,1074 @@
+{
+  "best_metric": 0.5945026874542236,
+  "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/FacebookAI/roberta_base_ledgar/checkpoint-2700",
+  "epoch": 2.878464818763326,
+  "eval_steps": 100,
+  "global_step": 2700,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "grad_norm": 3.73608136177063,
+      "learning_rate": 1.9822316986496094e-05,
+      "loss": 4.5378,
+      "step": 25
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 4.52000093460083,
+      "learning_rate": 1.9644633972992183e-05,
+      "loss": 4.1232,
+      "step": 50
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 6.089579105377197,
+      "learning_rate": 1.9466950959488275e-05,
+      "loss": 3.5699,
+      "step": 75
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 5.9484333992004395,
+      "learning_rate": 1.9289267945984364e-05,
+      "loss": 3.1362,
+      "step": 100
+    },
+    {
+      "epoch": 0.11,
+      "eval_accuracy": 0.5507,
+      "eval_f1_macro": 0.27432112841179934,
+      "eval_f1_micro": 0.5507,
+      "eval_loss": 2.7513325214385986,
+      "eval_runtime": 40.9521,
+      "eval_samples_per_second": 244.188,
+      "eval_steps_per_second": 3.834,
+      "step": 100
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 5.347239017486572,
+      "learning_rate": 1.9111584932480457e-05,
+      "loss": 2.7151,
+      "step": 125
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 6.90007209777832,
+      "learning_rate": 1.8933901918976546e-05,
+      "loss": 2.5145,
+      "step": 150
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 7.664882659912109,
+      "learning_rate": 1.8756218905472638e-05,
+      "loss": 2.1945,
+      "step": 175
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 6.626638889312744,
+      "learning_rate": 1.857853589196873e-05,
+      "loss": 2.0094,
+      "step": 200
+    },
+    {
+      "epoch": 0.21,
+      "eval_accuracy": 0.678,
+      "eval_f1_macro": 0.4120850985801981,
+      "eval_f1_micro": 0.678,
+      "eval_loss": 1.8194221258163452,
+      "eval_runtime": 40.6974,
+      "eval_samples_per_second": 245.716,
+      "eval_steps_per_second": 3.858,
+      "step": 200
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 5.4447102546691895,
+      "learning_rate": 1.8400852878464823e-05,
+      "loss": 1.8811,
+      "step": 225
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 7.736329555511475,
+      "learning_rate": 1.822316986496091e-05,
+      "loss": 1.7703,
+      "step": 250
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 7.9886322021484375,
+      "learning_rate": 1.8045486851457e-05,
+      "loss": 1.6891,
+      "step": 275
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 7.425679683685303,
+      "learning_rate": 1.7867803837953093e-05,
+      "loss": 1.573,
+      "step": 300
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.7322,
+      "eval_f1_macro": 0.5071623632300971,
+      "eval_f1_micro": 0.7322,
+      "eval_loss": 1.4166209697723389,
+      "eval_runtime": 40.5172,
+      "eval_samples_per_second": 246.809,
+      "eval_steps_per_second": 3.875,
+      "step": 300
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 11.348684310913086,
+      "learning_rate": 1.7690120824449185e-05,
+      "loss": 1.5468,
+      "step": 325
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 6.087052345275879,
+      "learning_rate": 1.7512437810945274e-05,
+      "loss": 1.441,
+      "step": 350
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 10.939698219299316,
+      "learning_rate": 1.7334754797441367e-05,
+      "loss": 1.3525,
+      "step": 375
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 5.3913373947143555,
+      "learning_rate": 1.715707178393746e-05,
+      "loss": 1.2789,
+      "step": 400
+    },
+    {
+      "epoch": 0.43,
+      "eval_accuracy": 0.7693,
+      "eval_f1_macro": 0.571119623062698,
+      "eval_f1_micro": 0.7693,
+      "eval_loss": 1.1782219409942627,
+      "eval_runtime": 40.4626,
+      "eval_samples_per_second": 247.142,
+      "eval_steps_per_second": 3.88,
+      "step": 400
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 5.381964683532715,
+      "learning_rate": 1.6979388770433548e-05,
+      "loss": 1.2842,
+      "step": 425
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 7.179779529571533,
+      "learning_rate": 1.6801705756929637e-05,
+      "loss": 1.3289,
+      "step": 450
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 6.146059513092041,
+      "learning_rate": 1.662402274342573e-05,
+      "loss": 1.2244,
+      "step": 475
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 9.716841697692871,
+      "learning_rate": 1.6446339729921822e-05,
+      "loss": 1.1898,
+      "step": 500
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.7877,
+      "eval_f1_macro": 0.6066845416751352,
+      "eval_f1_micro": 0.7877,
+      "eval_loss": 1.0304876565933228,
+      "eval_runtime": 40.4009,
+      "eval_samples_per_second": 247.519,
+      "eval_steps_per_second": 3.886,
+      "step": 500
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 6.480907440185547,
+      "learning_rate": 1.626865671641791e-05,
+      "loss": 1.1443,
+      "step": 525
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 5.845459461212158,
+      "learning_rate": 1.6090973702914003e-05,
+      "loss": 1.105,
+      "step": 550
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 6.848117351531982,
+      "learning_rate": 1.5913290689410095e-05,
+      "loss": 1.0441,
+      "step": 575
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 9.208276748657227,
+      "learning_rate": 1.5735607675906184e-05,
+      "loss": 1.0704,
+      "step": 600
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.7986,
+      "eval_f1_macro": 0.6314330168041138,
+      "eval_f1_micro": 0.7986,
+      "eval_loss": 0.9311466813087463,
+      "eval_runtime": 40.4137,
+      "eval_samples_per_second": 247.441,
+      "eval_steps_per_second": 3.885,
+      "step": 600
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 8.260689735412598,
+      "learning_rate": 1.5557924662402277e-05,
+      "loss": 1.0424,
+      "step": 625
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 7.713486671447754,
+      "learning_rate": 1.5380241648898366e-05,
+      "loss": 0.9894,
+      "step": 650
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 7.790108680725098,
+      "learning_rate": 1.5202558635394458e-05,
+      "loss": 0.996,
+      "step": 675
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 6.4116692543029785,
+      "learning_rate": 1.5024875621890549e-05,
+      "loss": 0.9539,
+      "step": 700
+    },
+    {
+      "epoch": 0.75,
+      "eval_accuracy": 0.8014,
+      "eval_f1_macro": 0.6331756093641667,
+      "eval_f1_micro": 0.8014,
+      "eval_loss": 0.8658352494239807,
+      "eval_runtime": 40.4285,
+      "eval_samples_per_second": 247.35,
+      "eval_steps_per_second": 3.883,
+      "step": 700
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 6.619067192077637,
+      "learning_rate": 1.484719260838664e-05,
+      "loss": 0.8869,
+      "step": 725
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 7.047544956207275,
+      "learning_rate": 1.4669509594882732e-05,
+      "loss": 0.9771,
+      "step": 750
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 7.623336315155029,
+      "learning_rate": 1.449182658137882e-05,
+      "loss": 0.9448,
+      "step": 775
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 6.41962194442749,
+      "learning_rate": 1.4314143567874911e-05,
+      "loss": 0.9228,
+      "step": 800
+    },
+    {
+      "epoch": 0.85,
+      "eval_accuracy": 0.8077,
+      "eval_f1_macro": 0.6462759356787142,
+      "eval_f1_micro": 0.8077,
+      "eval_loss": 0.8172683715820312,
+      "eval_runtime": 40.4547,
+      "eval_samples_per_second": 247.19,
+      "eval_steps_per_second": 3.881,
+      "step": 800
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 8.049590110778809,
+      "learning_rate": 1.4136460554371004e-05,
+      "loss": 0.9359,
+      "step": 825
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 6.472720623016357,
+      "learning_rate": 1.3958777540867094e-05,
+      "loss": 0.8501,
+      "step": 850
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 8.987987518310547,
+      "learning_rate": 1.3781094527363185e-05,
+      "loss": 0.8932,
+      "step": 875
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 7.016170024871826,
+      "learning_rate": 1.3603411513859277e-05,
+      "loss": 0.8447,
+      "step": 900
+    },
+    {
+      "epoch": 0.96,
+      "eval_accuracy": 0.8205,
+      "eval_f1_macro": 0.6760020998170815,
+      "eval_f1_micro": 0.8205,
+      "eval_loss": 0.7710947394371033,
+      "eval_runtime": 40.4462,
+      "eval_samples_per_second": 247.242,
+      "eval_steps_per_second": 3.882,
+      "step": 900
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 7.992467880249023,
+      "learning_rate": 1.3425728500355366e-05,
+      "loss": 0.8871,
+      "step": 925
+    },
+    {
+      "epoch": 1.01,
+      "grad_norm": 6.099947929382324,
+      "learning_rate": 1.3248045486851457e-05,
+      "loss": 0.8418,
+      "step": 950
+    },
+    {
+      "epoch": 1.04,
+      "grad_norm": 9.185840606689453,
+      "learning_rate": 1.307036247334755e-05,
+      "loss": 0.8124,
+      "step": 975
+    },
+    {
+      "epoch": 1.07,
+      "grad_norm": 6.065372467041016,
+      "learning_rate": 1.289267945984364e-05,
+      "loss": 0.7568,
+      "step": 1000
+    },
+    {
+      "epoch": 1.07,
+      "eval_accuracy": 0.8197,
+      "eval_f1_macro": 0.6749658745458524,
+      "eval_f1_micro": 0.8197,
+      "eval_loss": 0.753093957901001,
+      "eval_runtime": 40.4167,
+      "eval_samples_per_second": 247.422,
+      "eval_steps_per_second": 3.885,
+      "step": 1000
+    },
+    {
+      "epoch": 1.09,
+      "grad_norm": 5.7461981773376465,
+      "learning_rate": 1.271499644633973e-05,
+      "loss": 0.7704,
+      "step": 1025
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 10.816934585571289,
+      "learning_rate": 1.2537313432835823e-05,
+      "loss": 0.7348,
+      "step": 1050
+    },
+    {
+      "epoch": 1.15,
+      "grad_norm": 6.545167922973633,
+      "learning_rate": 1.2359630419331914e-05,
+      "loss": 0.7678,
+      "step": 1075
+    },
+    {
+      "epoch": 1.17,
+      "grad_norm": 9.452149391174316,
+      "learning_rate": 1.2181947405828003e-05,
+      "loss": 0.7484,
+      "step": 1100
+    },
+    {
+      "epoch": 1.17,
+      "eval_accuracy": 0.8219,
+      "eval_f1_macro": 0.682227300927334,
+      "eval_f1_micro": 0.8219,
+      "eval_loss": 0.730976402759552,
+      "eval_runtime": 40.4282,
+      "eval_samples_per_second": 247.352,
+      "eval_steps_per_second": 3.883,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 3.097867250442505,
+      "learning_rate": 1.2004264392324093e-05,
+      "loss": 0.7276,
+      "step": 1125
+    },
+    {
+      "epoch": 1.23,
+      "grad_norm": 5.687498092651367,
+      "learning_rate": 1.1826581378820186e-05,
+      "loss": 0.6838,
+      "step": 1150
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 9.114604949951172,
+      "learning_rate": 1.1648898365316276e-05,
+      "loss": 0.6807,
+      "step": 1175
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 7.552670955657959,
+      "learning_rate": 1.1471215351812369e-05,
+      "loss": 0.7915,
+      "step": 1200
+    },
+    {
+      "epoch": 1.28,
+      "eval_accuracy": 0.8271,
+      "eval_f1_macro": 0.6896865958100601,
+      "eval_f1_micro": 0.8271,
+      "eval_loss": 0.7162495851516724,
+      "eval_runtime": 40.4202,
+      "eval_samples_per_second": 247.401,
+      "eval_steps_per_second": 3.884,
+      "step": 1200
+    },
+    {
+      "epoch": 1.31,
+      "grad_norm": 8.283391952514648,
+      "learning_rate": 1.129353233830846e-05,
+      "loss": 0.7505,
+      "step": 1225
+    },
+    {
+      "epoch": 1.33,
+      "grad_norm": 4.6215434074401855,
+      "learning_rate": 1.111584932480455e-05,
+      "loss": 0.7226,
+      "step": 1250
+    },
+    {
+      "epoch": 1.36,
+      "grad_norm": 7.872697830200195,
+      "learning_rate": 1.0938166311300639e-05,
+      "loss": 0.7145,
+      "step": 1275
+    },
+    {
+      "epoch": 1.39,
+      "grad_norm": 6.955955982208252,
+      "learning_rate": 1.0760483297796731e-05,
+      "loss": 0.7083,
+      "step": 1300
+    },
+    {
+      "epoch": 1.39,
+      "eval_accuracy": 0.8256,
+      "eval_f1_macro": 0.6919298110503026,
+      "eval_f1_micro": 0.8256,
+      "eval_loss": 0.6928218603134155,
+      "eval_runtime": 40.4467,
+      "eval_samples_per_second": 247.239,
+      "eval_steps_per_second": 3.882,
+      "step": 1300
+    },
+    {
+      "epoch": 1.41,
+      "grad_norm": 8.991829872131348,
+      "learning_rate": 1.0582800284292822e-05,
+      "loss": 0.7001,
+      "step": 1325
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 5.560980796813965,
+      "learning_rate": 1.0405117270788913e-05,
+      "loss": 0.7247,
+      "step": 1350
+    },
+    {
+      "epoch": 1.47,
+      "grad_norm": 7.293112277984619,
+      "learning_rate": 1.0227434257285005e-05,
+      "loss": 0.6666,
+      "step": 1375
+    },
+    {
+      "epoch": 1.49,
+      "grad_norm": 6.286451816558838,
+      "learning_rate": 1.0049751243781096e-05,
+      "loss": 0.6292,
+      "step": 1400
+    },
+    {
+      "epoch": 1.49,
+      "eval_accuracy": 0.8281,
+      "eval_f1_macro": 0.6929689560154576,
+      "eval_f1_micro": 0.8281,
+      "eval_loss": 0.6823379397392273,
+      "eval_runtime": 40.4125,
+      "eval_samples_per_second": 247.448,
+      "eval_steps_per_second": 3.885,
+      "step": 1400
+    },
+    {
+      "epoch": 1.52,
+      "grad_norm": 8.192157745361328,
+      "learning_rate": 9.872068230277187e-06,
+      "loss": 0.707,
+      "step": 1425
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 6.905233383178711,
+      "learning_rate": 9.694385216773277e-06,
+      "loss": 0.7191,
+      "step": 1450
+    },
+    {
+      "epoch": 1.57,
+      "grad_norm": 7.920480251312256,
+      "learning_rate": 9.516702203269368e-06,
+      "loss": 0.7032,
+      "step": 1475
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 7.143226146697998,
+      "learning_rate": 9.339019189765458e-06,
+      "loss": 0.6646,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6,
+      "eval_accuracy": 0.8321,
+      "eval_f1_macro": 0.7047365437691051,
+      "eval_f1_micro": 0.8321,
+      "eval_loss": 0.6708748936653137,
+      "eval_runtime": 40.4198,
+      "eval_samples_per_second": 247.404,
+      "eval_steps_per_second": 3.884,
+      "step": 1500
+    },
+    {
+      "epoch": 1.63,
+      "grad_norm": 6.853357791900635,
+      "learning_rate": 9.161336176261551e-06,
+      "loss": 0.6678,
+      "step": 1525
+    },
+    {
+      "epoch": 1.65,
+      "grad_norm": 7.522273063659668,
+      "learning_rate": 8.98365316275764e-06,
+      "loss": 0.642,
+      "step": 1550
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 7.151234149932861,
+      "learning_rate": 8.805970149253732e-06,
+      "loss": 0.6268,
+      "step": 1575
+    },
+    {
+      "epoch": 1.71,
+      "grad_norm": 7.427074432373047,
+      "learning_rate": 8.628287135749823e-06,
+      "loss": 0.6808,
+      "step": 1600
+    },
+    {
+      "epoch": 1.71,
+      "eval_accuracy": 0.8348,
+      "eval_f1_macro": 0.7160389397092434,
+      "eval_f1_micro": 0.8348,
+      "eval_loss": 0.6589647531509399,
+      "eval_runtime": 40.4351,
+      "eval_samples_per_second": 247.31,
+      "eval_steps_per_second": 3.883,
+      "step": 1600
+    },
+    {
+      "epoch": 1.73,
+      "grad_norm": 6.88338565826416,
+      "learning_rate": 8.450604122245915e-06,
+      "loss": 0.6485,
+      "step": 1625
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 6.0003228187561035,
+      "learning_rate": 8.272921108742004e-06,
+      "loss": 0.6559,
+      "step": 1650
+    },
+    {
+      "epoch": 1.79,
+      "grad_norm": 7.314059734344482,
+      "learning_rate": 8.095238095238097e-06,
+      "loss": 0.6229,
+      "step": 1675
+    },
+    {
+      "epoch": 1.81,
+      "grad_norm": 7.924875736236572,
+      "learning_rate": 7.917555081734187e-06,
+      "loss": 0.6748,
+      "step": 1700
+    },
+    {
+      "epoch": 1.81,
+      "eval_accuracy": 0.8336,
+      "eval_f1_macro": 0.7130021255940269,
+      "eval_f1_micro": 0.8336,
+      "eval_loss": 0.6455573439598083,
+      "eval_runtime": 40.4285,
+      "eval_samples_per_second": 247.351,
+      "eval_steps_per_second": 3.883,
+      "step": 1700
+    },
+    {
+      "epoch": 1.84,
+      "grad_norm": 13.616132736206055,
+      "learning_rate": 7.739872068230278e-06,
+      "loss": 0.7141,
+      "step": 1725
+    },
+    {
+      "epoch": 1.87,
+      "grad_norm": 8.604755401611328,
+      "learning_rate": 7.5621890547263685e-06,
+      "loss": 0.6234,
+      "step": 1750
+    },
+    {
+      "epoch": 1.89,
+      "grad_norm": 10.16673469543457,
+      "learning_rate": 7.38450604122246e-06,
+      "loss": 0.6856,
+      "step": 1775
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 10.532857894897461,
+      "learning_rate": 7.20682302771855e-06,
+      "loss": 0.6251,
+      "step": 1800
+    },
+    {
+      "epoch": 1.92,
+      "eval_accuracy": 0.8377,
+      "eval_f1_macro": 0.7166521934010971,
+      "eval_f1_micro": 0.8377,
+      "eval_loss": 0.6294605731964111,
+      "eval_runtime": 40.4382,
+      "eval_samples_per_second": 247.291,
+      "eval_steps_per_second": 3.882,
+      "step": 1800
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 7.858551025390625,
+      "learning_rate": 7.029140014214641e-06,
+      "loss": 0.7219,
+      "step": 1825
+    },
+    {
+      "epoch": 1.97,
+      "grad_norm": 10.671606063842773,
+      "learning_rate": 6.851457000710733e-06,
+      "loss": 0.6626,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 8.557258605957031,
+      "learning_rate": 6.673773987206824e-06,
+      "loss": 0.6049,
+      "step": 1875
+    },
+    {
+      "epoch": 2.03,
+      "grad_norm": 5.789582252502441,
+      "learning_rate": 6.496090973702914e-06,
+      "loss": 0.6155,
+      "step": 1900
+    },
+    {
+      "epoch": 2.03,
+      "eval_accuracy": 0.8382,
+      "eval_f1_macro": 0.7147165247941616,
+      "eval_f1_micro": 0.8382,
+      "eval_loss": 0.6280255913734436,
+      "eval_runtime": 40.4506,
+      "eval_samples_per_second": 247.215,
+      "eval_steps_per_second": 3.881,
+      "step": 1900
+    },
+    {
+      "epoch": 2.05,
+      "grad_norm": 7.670330047607422,
+      "learning_rate": 6.318407960199006e-06,
+      "loss": 0.5902,
+      "step": 1925
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 7.693389892578125,
+      "learning_rate": 6.140724946695097e-06,
+      "loss": 0.5899,
+      "step": 1950
+    },
+    {
+      "epoch": 2.11,
+      "grad_norm": 5.274115562438965,
+      "learning_rate": 5.963041933191187e-06,
+      "loss": 0.5834,
+      "step": 1975
+    },
+    {
+      "epoch": 2.13,
+      "grad_norm": 6.170833587646484,
+      "learning_rate": 5.785358919687279e-06,
+      "loss": 0.5387,
+      "step": 2000
+    },
+    {
+      "epoch": 2.13,
+      "eval_accuracy": 0.8387,
+      "eval_f1_macro": 0.7212820719554683,
+      "eval_f1_micro": 0.8387,
+      "eval_loss": 0.6256607174873352,
+      "eval_runtime": 40.4329,
+      "eval_samples_per_second": 247.324,
+      "eval_steps_per_second": 3.883,
+      "step": 2000
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 8.71805477142334,
+      "learning_rate": 5.607675906183369e-06,
+      "loss": 0.5793,
+      "step": 2025
+    },
+    {
+      "epoch": 2.19,
+      "grad_norm": 5.482748985290527,
+      "learning_rate": 5.42999289267946e-06,
+      "loss": 0.5901,
+      "step": 2050
+    },
+    {
+      "epoch": 2.21,
+      "grad_norm": 9.235983848571777,
+      "learning_rate": 5.2523098791755514e-06,
+      "loss": 0.5602,
+      "step": 2075
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 9.202109336853027,
+      "learning_rate": 5.074626865671642e-06,
+      "loss": 0.5859,
+      "step": 2100
+    },
+    {
+      "epoch": 2.24,
+      "eval_accuracy": 0.8402,
+      "eval_f1_macro": 0.7276460225327179,
+      "eval_f1_micro": 0.8402,
+      "eval_loss": 0.61931312084198,
+      "eval_runtime": 40.4383,
+      "eval_samples_per_second": 247.29,
+      "eval_steps_per_second": 3.882,
+      "step": 2100
+    },
+    {
+      "epoch": 2.27,
+      "grad_norm": 8.1707124710083,
+      "learning_rate": 4.896943852167733e-06,
+      "loss": 0.5463,
+      "step": 2125
+    },
+    {
+      "epoch": 2.29,
+      "grad_norm": 10.29979133605957,
+      "learning_rate": 4.719260838663824e-06,
+      "loss": 0.5454,
+      "step": 2150
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 10.924964904785156,
+      "learning_rate": 4.541577825159915e-06,
+      "loss": 0.5126,
+      "step": 2175
+    },
+    {
+      "epoch": 2.35,
+      "grad_norm": 8.300230979919434,
+      "learning_rate": 4.363894811656006e-06,
+      "loss": 0.5663,
+      "step": 2200
+    },
+    {
+      "epoch": 2.35,
+      "eval_accuracy": 0.843,
+      "eval_f1_macro": 0.7314175321043476,
+      "eval_f1_micro": 0.843,
+      "eval_loss": 0.6136118173599243,
+      "eval_runtime": 40.4132,
+      "eval_samples_per_second": 247.444,
+      "eval_steps_per_second": 3.885,
+      "step": 2200
+    },
+    {
+      "epoch": 2.37,
+      "grad_norm": 6.390136241912842,
+      "learning_rate": 4.186211798152097e-06,
+      "loss": 0.5346,
+      "step": 2225
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 8.411725997924805,
+      "learning_rate": 4.008528784648188e-06,
+      "loss": 0.5544,
+      "step": 2250
+    },
+    {
+      "epoch": 2.43,
+      "grad_norm": 7.073245525360107,
+      "learning_rate": 3.8308457711442784e-06,
+      "loss": 0.5668,
+      "step": 2275
+    },
+    {
+      "epoch": 2.45,
+      "grad_norm": 10.046344757080078,
+      "learning_rate": 3.65316275764037e-06,
+      "loss": 0.5836,
+      "step": 2300
+    },
+    {
+      "epoch": 2.45,
+      "eval_accuracy": 0.8433,
+      "eval_f1_macro": 0.7320187690833233,
+      "eval_f1_micro": 0.8433,
+      "eval_loss": 0.6109758615493774,
+      "eval_runtime": 40.4128,
+      "eval_samples_per_second": 247.446,
+      "eval_steps_per_second": 3.885,
+      "step": 2300
+    },
+    {
+      "epoch": 2.48,
+      "grad_norm": 8.907343864440918,
+      "learning_rate": 3.4754797441364606e-06,
+      "loss": 0.5855,
+      "step": 2325
+    },
+    {
+      "epoch": 2.51,
+      "grad_norm": 12.943143844604492,
+      "learning_rate": 3.297796730632552e-06,
+      "loss": 0.5892,
+      "step": 2350
+    },
+    {
+      "epoch": 2.53,
+      "grad_norm": 9.387904167175293,
+      "learning_rate": 3.120113717128643e-06,
+      "loss": 0.5446,
+      "step": 2375
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 6.5862650871276855,
+      "learning_rate": 2.9424307036247335e-06,
+      "loss": 0.5451,
+      "step": 2400
+    },
+    {
+      "epoch": 2.56,
+      "eval_accuracy": 0.8463,
+      "eval_f1_macro": 0.7385431611673754,
+      "eval_f1_micro": 0.8463,
+      "eval_loss": 0.5996297001838684,
+      "eval_runtime": 40.4138,
+      "eval_samples_per_second": 247.44,
+      "eval_steps_per_second": 3.885,
+      "step": 2400
+    },
+    {
+      "epoch": 2.59,
+      "grad_norm": 5.578644275665283,
+      "learning_rate": 2.764747690120825e-06,
+      "loss": 0.6264,
+      "step": 2425
+    },
+    {
+      "epoch": 2.61,
+      "grad_norm": 8.960025787353516,
+      "learning_rate": 2.5870646766169156e-06,
+      "loss": 0.578,
+      "step": 2450
+    },
+    {
+      "epoch": 2.64,
+      "grad_norm": 9.928258895874023,
+      "learning_rate": 2.4093816631130067e-06,
+      "loss": 0.5511,
+      "step": 2475
+    },
+    {
+      "epoch": 2.67,
+      "grad_norm": 10.85412883758545,
+      "learning_rate": 2.2316986496090974e-06,
+      "loss": 0.5198,
+      "step": 2500
+    },
+    {
+      "epoch": 2.67,
+      "eval_accuracy": 0.8445,
+      "eval_f1_macro": 0.738451468801292,
+      "eval_f1_micro": 0.8445,
+      "eval_loss": 0.5995715856552124,
+      "eval_runtime": 40.4135,
+      "eval_samples_per_second": 247.442,
+      "eval_steps_per_second": 3.885,
+      "step": 2500
+    },
+    {
+      "epoch": 2.69,
+      "grad_norm": 5.94683313369751,
+      "learning_rate": 2.0540156361051885e-06,
+      "loss": 0.5483,
+      "step": 2525
+    },
+    {
+      "epoch": 2.72,
+      "grad_norm": 6.956961154937744,
+      "learning_rate": 1.8763326226012796e-06,
+      "loss": 0.5464,
+      "step": 2550
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 11.097511291503906,
+      "learning_rate": 1.6986496090973702e-06,
+      "loss": 0.5505,
+      "step": 2575
+    },
+    {
+      "epoch": 2.77,
+      "grad_norm": 11.86330795288086,
+      "learning_rate": 1.5209665955934613e-06,
+      "loss": 0.5508,
+      "step": 2600
+    },
+    {
+      "epoch": 2.77,
+      "eval_accuracy": 0.8439,
+      "eval_f1_macro": 0.7371498369473302,
+      "eval_f1_micro": 0.8439,
+      "eval_loss": 0.5972757935523987,
+      "eval_runtime": 40.4468,
+      "eval_samples_per_second": 247.238,
+      "eval_steps_per_second": 3.882,
+      "step": 2600
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 7.698044776916504,
+      "learning_rate": 1.3432835820895524e-06,
+      "loss": 0.5485,
+      "step": 2625
+    },
+    {
+      "epoch": 2.83,
+      "grad_norm": 8.738574981689453,
+      "learning_rate": 1.1656005685856433e-06,
+      "loss": 0.6119,
+      "step": 2650
+    },
+    {
+      "epoch": 2.85,
+      "grad_norm": 5.974484443664551,
+      "learning_rate": 9.879175550817342e-07,
+      "loss": 0.5344,
+      "step": 2675
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 6.997408866882324,
+      "learning_rate": 8.102345415778253e-07,
+      "loss": 0.5199,
+      "step": 2700
+    },
+    {
+      "epoch": 2.88,
+      "eval_accuracy": 0.8453,
+      "eval_f1_macro": 0.7391787821862769,
+      "eval_f1_micro": 0.8453,
+      "eval_loss": 0.5945026874542236,
+      "eval_runtime": 40.443,
+      "eval_samples_per_second": 247.261,
+      "eval_steps_per_second": 3.882,
+      "step": 2700
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 2814,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 4.550559511294771e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/checkpoint-2700/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7aceee0724347879f40ba735916b878d3344cbcfbb26ba8f61146b41664c76a
+size 5048

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/config.json

@@ -0,0 +1,233 @@
+{
+  "_name_or_path": "FacebookAI/roberta-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "text-classification",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "0",
+    "1": "1",
+    "2": "10",
+    "3": "11",
+    "4": "12",
+    "5": "13",
+    "6": "14",
+    "7": "15",
+    "8": "16",
+    "9": "17",
+    "10": "18",
+    "11": "19",
+    "12": "2",
+    "13": "20",
+    "14": "21",
+    "15": "22",
+    "16": "23",
+    "17": "24",
+    "18": "25",
+    "19": "26",
+    "20": "27",
+    "21": "28",
+    "22": "29",
+    "23": "3",
+    "24": "30",
+    "25": "31",
+    "26": "32",
+    "27": "33",
+    "28": "34",
+    "29": "35",
+    "30": "36",
+    "31": "37",
+    "32": "38",
+    "33": "39",
+    "34": "4",
+    "35": "40",
+    "36": "41",
+    "37": "42",
+    "38": "43",
+    "39": "44",
+    "40": "45",
+    "41": "46",
+    "42": "47",
+    "43": "48",
+    "44": "49",
+    "45": "5",
+    "46": "50",
+    "47": "51",
+    "48": "52",
+    "49": "53",
+    "50": "54",
+    "51": "55",
+    "52": "56",
+    "53": "57",
+    "54": "58",
+    "55": "59",
+    "56": "6",
+    "57": "60",
+    "58": "61",
+    "59": "62",
+    "60": "63",
+    "61": "64",
+    "62": "65",
+    "63": "66",
+    "64": "67",
+    "65": "68",
+    "66": "69",
+    "67": "7",
+    "68": "70",
+    "69": "71",
+    "70": "72",
+    "71": "73",
+    "72": "74",
+    "73": "75",
+    "74": "76",
+    "75": "77",
+    "76": "78",
+    "77": "79",
+    "78": "8",
+    "79": "80",
+    "80": "81",
+    "81": "82",
+    "82": "83",
+    "83": "84",
+    "84": "85",
+    "85": "86",
+    "86": "87",
+    "87": "88",
+    "88": "89",
+    "89": "9",
+    "90": "90",
+    "91": "91",
+    "92": "92",
+    "93": "93",
+    "94": "94",
+    "95": "95",
+    "96": "96",
+    "97": "97",
+    "98": "98",
+    "99": "99"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "0": 0,
+    "1": 1,
+    "10": 2,
+    "11": 3,
+    "12": 4,
+    "13": 5,
+    "14": 6,
+    "15": 7,
+    "16": 8,
+    "17": 9,
+    "18": 10,
+    "19": 11,
+    "2": 12,
+    "20": 13,
+    "21": 14,
+    "22": 15,
+    "23": 16,
+    "24": 17,
+    "25": 18,
+    "26": 19,
+    "27": 20,
+    "28": 21,
+    "29": 22,
+    "3": 23,
+    "30": 24,
+    "31": 25,
+    "32": 26,
+    "33": 27,
+    "34": 28,
+    "35": 29,
+    "36": 30,
+    "37": 31,
+    "38": 32,
+    "39": 33,
+    "4": 34,
+    "40": 35,
+    "41": 36,
+    "42": 37,
+    "43": 38,
+    "44": 39,
+    "45": 40,
+    "46": 41,
+    "47": 42,
+    "48": 43,
+    "49": 44,
+    "5": 45,
+    "50": 46,
+    "51": 47,
+    "52": 48,
+    "53": 49,
+    "54": 50,
+    "55": 51,
+    "56": 52,
+    "57": 53,
+    "58": 54,
+    "59": 55,
+    "6": 56,
+    "60": 57,
+    "61": 58,
+    "62": 59,
+    "63": 60,
+    "64": 61,
+    "65": 62,
+    "66": 63,
+    "67": 64,
+    "68": 65,
+    "69": 66,
+    "7": 67,
+    "70": 68,
+    "71": 69,
+    "72": 70,
+    "73": 71,
+    "74": 72,
+    "75": 73,
+    "76": 74,
+    "77": 75,
+    "78": 76,
+    "79": 77,
+    "8": 78,
+    "80": 79,
+    "81": 80,
+    "82": 81,
+    "83": 82,
+    "84": 83,
+    "85": 84,
+    "86": 85,
+    "87": 86,
+    "88": 87,
+    "89": 88,
+    "9": 89,
+    "90": 90,
+    "91": 91,
+    "92": 92,
+    "93": 93,
+    "94": 94,
+    "95": 95,
+    "96": 96,
+    "97": 97,
+    "98": 98,
+    "99": 99
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8453,
+    "eval_f1_macro": 0.7391787821862769,
+    "eval_f1_micro": 0.8453,
+    "eval_loss": 0.5945026874542236,
+    "eval_runtime": 40.1416,
+    "eval_samples": 10000,
+    "eval_samples_per_second": 249.118,
+    "eval_steps_per_second": 3.911
+}

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24009191e7c7cf88a3be5f0d6c77255d3e41fe6b0bd9186ec387ce66d42c73ad
+size 498914272

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/run.log

@@ -0,0 +1,4 @@
+03/21/2024 06:26:13 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, 16-bits training: False
+03/21/2024 06:26:13 - WARNING - __main__ - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, 16-bits training: False
+03/21/2024 06:26:16 - WARNING - __main__ - The label2id key in the model config.json is not equal to the label2id key of this run. You can ignore this if you are doing finetuning.
+03/21/2024 06:26:16 - WARNING - __main__ - The label2id key in the model config.json is not equal to the label2id key of this run. You can ignore this if you are doing finetuning.

max_seq_length_512_experiments/FacebookAI/roberta_base_ledgar/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>"
+}