Initial commit

README.md

@@ -1,3 +1,88 @@
 ---
-license: apache-2.0
+license: mit
+tags:
+- generated_from_trainer
+datasets:
+- it5/datasets
+metrics:
+- rouge
+model-index:
+- name: it5-efficient-small-el32-st_r2g-0.0003
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: it5/datasets st_r2g
+      type: it5/datasets
+      args: st_r2g
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 30.0502
 ---
+
+<!-- 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. -->
+
+# it5-efficient-small-el32-st_r2g-0.0003
+
+This model is a fine-tuned version of [stefan-it/it5-efficient-small-el32](https://huggingface.co/stefan-it/it5-efficient-small-el32) on the it5/datasets st_r2g dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.6135
+- Rouge1: 30.0502
+- Rouge2: 11.5687
+- Rougel: 26.5953
+- Rougelsum: 27.0402
+- Gen Len: 16.9578
+
+## 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: 0.0003
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
+|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
+| 3.1265        | 0.74  | 5000  | 2.7247          | 26.8378 | 9.3464  | 23.9521 | 24.2837   | 15.5914 |
+| 2.8786        | 1.49  | 10000 | 2.6532          | 27.5869 | 10.0861 | 24.7406 | 25.0245   | 15.3272 |
+| 2.6587        | 2.23  | 15000 | 2.6080          | 28.2336 | 10.5229 | 25.3053 | 25.6716   | 15.4338 |
+| 2.664         | 2.98  | 20000 | 2.5630          | 28.6673 | 10.8421 | 25.7032 | 26.0245   | 15.6255 |
+| 2.4896        | 3.72  | 25000 | 2.5679          | 28.842  | 10.885  | 25.6757 | 26.0633   | 16.1841 |
+| 2.34          | 4.47  | 30000 | 2.5564          | 29.3246 | 11.1981 | 26.1637 | 26.5392   | 15.7826 |
+| 2.2204        | 5.21  | 35000 | 2.5744          | 29.5545 | 11.3806 | 26.3237 | 26.6993   | 15.8374 |
+| 2.2301        | 5.96  | 40000 | 2.5614          | 29.5872 | 11.4227 | 26.3139 | 26.7196   | 15.7213 |
+| 2.1219        | 6.7   | 45000 | 2.5617          | 29.8256 | 11.3702 | 26.4156 | 26.8465   | 15.936  |
+| 2.007         | 7.45  | 50000 | 2.6014          | 29.743  | 11.4336 | 26.38   | 26.772    | 15.7144 |
+| 1.9398        | 8.19  | 55000 | 2.6080          | 29.9478 | 11.4801 | 26.5352 | 26.9746   | 15.9308 |
+| 1.9426        | 8.94  | 60000 | 2.6022          | 30.097  | 11.5602 | 26.705  | 27.1092   | 15.8598 |
+| 1.8853        | 9.68  | 65000 | 2.6138          | 30.1588 | 11.5823 | 26.6984 | 27.1371   | 15.803  |
+
+
+### Framework versions
+
+- Transformers 4.15.0
+- Pytorch 1.10.0+cu102
+- Datasets 1.17.0
+- Tokenizers 0.10.3

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 10.0,
+    "eval_gen_len": 16.9578,
+    "eval_loss": 2.6134610176086426,
+    "eval_rouge1": 30.0502,
+    "eval_rouge2": 11.5687,
+    "eval_rougeL": 26.5953,
+    "eval_rougeLsum": 27.0402,
+    "eval_runtime": 582.0642,
+    "eval_samples": 10000,
+    "eval_samples_per_second": 17.18,
+    "eval_steps_per_second": 2.148,
+    "train_loss": 2.3876063485090873,
+    "train_runtime": 39696.1157,
+    "train_samples": 53701,
+    "train_samples_per_second": 13.528,
+    "train_steps_per_second": 1.691
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": ".",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "d_ff": 2048,
+  "d_kv": 64,
+  "d_model": 512,
+  "decoder_start_token_id": 0,
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "relu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "n_positions": 512,
+  "num_decoder_layers": 6,
+  "num_heads": 8,
+  "num_layers": 32,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "torch_dtype": "float32",
+  "transformers_version": "4.15.0",
+  "use_cache": true,
+  "vocab_size": 32100
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 10.0,
+    "eval_gen_len": 16.9578,
+    "eval_loss": 2.6134610176086426,
+    "eval_rouge1": 30.0502,
+    "eval_rouge2": 11.5687,
+    "eval_rougeL": 26.5953,
+    "eval_rougeLsum": 27.0402,
+    "eval_runtime": 582.0642,
+    "eval_samples": 10000,
+    "eval_samples_per_second": 17.18,
+    "eval_steps_per_second": 2.148
+}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f32f90aca865f75083753513b2b24474e246018fdbeeb3bc535db1c5be83edf5
+size 569246164

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6efe540637765018901b6e84d1ac7e5b5607937125de42cd0c9532f3ef1858f
+size 569387035

runs/Apr28_20-05-28_pg-gpu26/1651169174.545433/events.out.tfevents.1651169174.pg-gpu26.22885.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:920b4fd18410a4c0e175a4ebc37de893944eee6ca540d12ae522a036b2e0c9f7
+size 5139

runs/Apr28_20-05-28_pg-gpu26/events.out.tfevents.1651169174.pg-gpu26.22885.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3098766f0e1c89dfd596e1d071d3fb4baa4dadf3e53c794664fc65fadf20e8c6
+size 32267

runs/Apr28_20-05-28_pg-gpu26/events.out.tfevents.1651209455.pg-gpu26.22885.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:954a9731b77ac33087e982339d55d0df962d750b70255902ef6092b831b40d3e
+size 575

special_tokens_map.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dffd01fc009b7e92d98eddff8853983e271b41302ed0d363000e8581df12000
+size 817200

tf_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4d6d87fd311c59ea51057e9f43ebe541e155cf587b4fb3806011b806897348e
+size 569947488

tokenizer_config.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "special_tokens_map_file": null, "name_or_path": "stefan-it/it5-efficient-small-el32", "sp_model_kwargs": {}, "tokenizer_class": "T5Tokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 2.3876063485090873,
+    "train_runtime": 39696.1157,
+    "train_samples": 53701,
+    "train_samples_per_second": 13.528,
+    "train_steps_per_second": 1.691
+}

trainer_state.json

@@ -0,0 +1,998 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "global_step": 67130,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.000297765529569492,
+      "loss": 4.1249,
+      "step": 500
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029553105913898405,
+      "loss": 3.6267,
+      "step": 1000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00029329658870847603,
+      "loss": 3.4522,
+      "step": 1500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00029106211827796807,
+      "loss": 3.3665,
+      "step": 2000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002888276478474601,
+      "loss": 3.3154,
+      "step": 2500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0002865931774169522,
+      "loss": 3.2553,
+      "step": 3000
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0002843587069864442,
+      "loss": 3.2225,
+      "step": 3500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002821242365559362,
+      "loss": 3.1854,
+      "step": 4000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00027988976612542825,
+      "loss": 3.1867,
+      "step": 4500
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.0002776552956949203,
+      "loss": 3.1265,
+      "step": 5000
+    },
+    {
+      "epoch": 0.74,
+      "eval_gen_len": 15.5914,
+      "eval_loss": 2.7246766090393066,
+      "eval_rouge1": 26.8378,
+      "eval_rouge2": 9.3464,
+      "eval_rougeL": 23.9521,
+      "eval_rougeLsum": 24.2837,
+      "eval_runtime": 527.2879,
+      "eval_samples_per_second": 18.965,
+      "eval_steps_per_second": 2.371,
+      "step": 5000
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002754208252644123,
+      "loss": 3.1287,
+      "step": 5500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0002731863548339043,
+      "loss": 3.1078,
+      "step": 6000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00027095188440339634,
+      "loss": 3.0787,
+      "step": 6500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0002687174139728884,
+      "loss": 3.0054,
+      "step": 7000
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0002664829435423804,
+      "loss": 2.8889,
+      "step": 7500
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00026424847311187245,
+      "loss": 2.9183,
+      "step": 8000
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0002620140026813645,
+      "loss": 2.8848,
+      "step": 8500
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00025977953225085653,
+      "loss": 2.872,
+      "step": 9000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00025754506182034857,
+      "loss": 2.898,
+      "step": 9500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.0002553105913898406,
+      "loss": 2.8786,
+      "step": 10000
+    },
+    {
+      "epoch": 1.49,
+      "eval_gen_len": 15.3272,
+      "eval_loss": 2.653170347213745,
+      "eval_rouge1": 27.5869,
+      "eval_rouge2": 10.0861,
+      "eval_rougeL": 24.7406,
+      "eval_rougeLsum": 25.0245,
+      "eval_runtime": 532.273,
+      "eval_samples_per_second": 18.787,
+      "eval_steps_per_second": 2.348,
+      "step": 10000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0002530761209593326,
+      "loss": 2.8664,
+      "step": 10500
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.0002508416505288246,
+      "loss": 2.873,
+      "step": 11000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.00024860718009831666,
+      "loss": 2.8685,
+      "step": 11500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.0002463727096678087,
+      "loss": 2.8713,
+      "step": 12000
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00024413823923730073,
+      "loss": 2.8558,
+      "step": 12500
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00024190376880679277,
+      "loss": 2.8361,
+      "step": 13000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.0002396692983762848,
+      "loss": 2.8026,
+      "step": 13500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.00023743482794577684,
+      "loss": 2.6527,
+      "step": 14000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 0.00023520035751526885,
+      "loss": 2.6452,
+      "step": 14500
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.0002329658870847609,
+      "loss": 2.6587,
+      "step": 15000
+    },
+    {
+      "epoch": 2.23,
+      "eval_gen_len": 15.4338,
+      "eval_loss": 2.6080288887023926,
+      "eval_rouge1": 28.2336,
+      "eval_rouge2": 10.5229,
+      "eval_rougeL": 25.3053,
+      "eval_rougeLsum": 25.6716,
+      "eval_runtime": 535.3754,
+      "eval_samples_per_second": 18.678,
+      "eval_steps_per_second": 2.335,
+      "step": 15000
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.00023073141665425293,
+      "loss": 2.6667,
+      "step": 15500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 0.00022849694622374496,
+      "loss": 2.6723,
+      "step": 16000
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.000226262475793237,
+      "loss": 2.6652,
+      "step": 16500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.000224028005362729,
+      "loss": 2.6577,
+      "step": 17000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.00022179353493222105,
+      "loss": 2.6673,
+      "step": 17500
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00021955906450171308,
+      "loss": 2.6533,
+      "step": 18000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.00021732459407120512,
+      "loss": 2.6571,
+      "step": 18500
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.00021509012364069713,
+      "loss": 2.6636,
+      "step": 19000
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.00021285565321018917,
+      "loss": 2.6579,
+      "step": 19500
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 0.0002106211827796812,
+      "loss": 2.664,
+      "step": 20000
+    },
+    {
+      "epoch": 2.98,
+      "eval_gen_len": 15.6255,
+      "eval_loss": 2.5629758834838867,
+      "eval_rouge1": 28.6673,
+      "eval_rouge2": 10.8421,
+      "eval_rougeL": 25.7032,
+      "eval_rougeLsum": 26.0245,
+      "eval_runtime": 536.8291,
+      "eval_samples_per_second": 18.628,
+      "eval_steps_per_second": 2.328,
+      "step": 20000
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 0.00020838671234917324,
+      "loss": 2.512,
+      "step": 20500
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 0.00020615224191866528,
+      "loss": 2.4748,
+      "step": 21000
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 0.0002039177714881573,
+      "loss": 2.4773,
+      "step": 21500
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 0.00020168330105764932,
+      "loss": 2.4694,
+      "step": 22000
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 0.00019944883062714136,
+      "loss": 2.4928,
+      "step": 22500
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 0.0001972143601966334,
+      "loss": 2.4941,
+      "step": 23000
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 0.0001949798897661254,
+      "loss": 2.4867,
+      "step": 23500
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 0.00019274541933561744,
+      "loss": 2.4978,
+      "step": 24000
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 0.00019051094890510948,
+      "loss": 2.5118,
+      "step": 24500
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 0.00018827647847460152,
+      "loss": 2.4896,
+      "step": 25000
+    },
+    {
+      "epoch": 3.72,
+      "eval_gen_len": 16.1841,
+      "eval_loss": 2.567906141281128,
+      "eval_rouge1": 28.842,
+      "eval_rouge2": 10.885,
+      "eval_rougeL": 25.6757,
+      "eval_rougeLsum": 26.0633,
+      "eval_runtime": 528.6739,
+      "eval_samples_per_second": 18.915,
+      "eval_steps_per_second": 2.364,
+      "step": 25000
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 0.00018604200804409355,
+      "loss": 2.5141,
+      "step": 25500
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 0.00018380753761358556,
+      "loss": 2.5001,
+      "step": 26000
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 0.0001815730671830776,
+      "loss": 2.4924,
+      "step": 26500
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 0.00017933859675256964,
+      "loss": 2.4512,
+      "step": 27000
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 0.00017710412632206167,
+      "loss": 2.3092,
+      "step": 27500
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 0.00017486965589155368,
+      "loss": 2.3359,
+      "step": 28000
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 0.00017263518546104572,
+      "loss": 2.3522,
+      "step": 28500
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 0.00017040071503053776,
+      "loss": 2.3723,
+      "step": 29000
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 0.0001681662446000298,
+      "loss": 2.3489,
+      "step": 29500
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 0.00016593177416952183,
+      "loss": 2.34,
+      "step": 30000
+    },
+    {
+      "epoch": 4.47,
+      "eval_gen_len": 15.7826,
+      "eval_loss": 2.5564002990722656,
+      "eval_rouge1": 29.3246,
+      "eval_rouge2": 11.1981,
+      "eval_rougeL": 26.1637,
+      "eval_rougeLsum": 26.5392,
+      "eval_runtime": 536.1351,
+      "eval_samples_per_second": 18.652,
+      "eval_steps_per_second": 2.332,
+      "step": 30000
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 0.00016369730373901384,
+      "loss": 2.3463,
+      "step": 30500
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 0.00016146283330850588,
+      "loss": 2.3537,
+      "step": 31000
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 0.00015922836287799791,
+      "loss": 2.345,
+      "step": 31500
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 0.00015699389244748995,
+      "loss": 2.345,
+      "step": 32000
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 0.00015475942201698196,
+      "loss": 2.3586,
+      "step": 32500
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 0.000152524951586474,
+      "loss": 2.3495,
+      "step": 33000
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 0.00015029048115596604,
+      "loss": 2.3716,
+      "step": 33500
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 0.00014805601072545804,
+      "loss": 2.216,
+      "step": 34000
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 0.00014582154029495008,
+      "loss": 2.211,
+      "step": 34500
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 0.00014358706986444212,
+      "loss": 2.2204,
+      "step": 35000
+    },
+    {
+      "epoch": 5.21,
+      "eval_gen_len": 15.8374,
+      "eval_loss": 2.5744452476501465,
+      "eval_rouge1": 29.5545,
+      "eval_rouge2": 11.3806,
+      "eval_rougeL": 26.3237,
+      "eval_rougeLsum": 26.6993,
+      "eval_runtime": 531.1942,
+      "eval_samples_per_second": 18.826,
+      "eval_steps_per_second": 2.353,
+      "step": 35000
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 0.00014135259943393416,
+      "loss": 2.2043,
+      "step": 35500
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 0.0001391181290034262,
+      "loss": 2.2349,
+      "step": 36000
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 0.0001368836585729182,
+      "loss": 2.2155,
+      "step": 36500
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 0.00013464918814241024,
+      "loss": 2.2326,
+      "step": 37000
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 0.00013241471771190228,
+      "loss": 2.2205,
+      "step": 37500
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 0.0001301802472813943,
+      "loss": 2.2288,
+      "step": 38000
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 0.00012794577685088632,
+      "loss": 2.228,
+      "step": 38500
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 0.00012571130642037836,
+      "loss": 2.2365,
+      "step": 39000
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 0.00012347683598987037,
+      "loss": 2.2164,
+      "step": 39500
+    },
+    {
+      "epoch": 5.96,
+      "learning_rate": 0.00012124236555936242,
+      "loss": 2.2301,
+      "step": 40000
+    },
+    {
+      "epoch": 5.96,
+      "eval_gen_len": 15.7213,
+      "eval_loss": 2.5613908767700195,
+      "eval_rouge1": 29.5872,
+      "eval_rouge2": 11.4227,
+      "eval_rougeL": 26.3139,
+      "eval_rougeLsum": 26.7196,
+      "eval_runtime": 535.5579,
+      "eval_samples_per_second": 18.672,
+      "eval_steps_per_second": 2.334,
+      "step": 40000
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 0.00011900789512885446,
+      "loss": 2.1538,
+      "step": 40500
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 0.00011677342469834648,
+      "loss": 2.0981,
+      "step": 41000
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 0.00011453895426783852,
+      "loss": 2.0973,
+      "step": 41500
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 0.00011230448383733054,
+      "loss": 2.0959,
+      "step": 42000
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 0.00011007001340682258,
+      "loss": 2.1013,
+      "step": 42500
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 0.0001078355429763146,
+      "loss": 2.1126,
+      "step": 43000
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 0.00010560107254580664,
+      "loss": 2.097,
+      "step": 43500
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 0.00010336660211529866,
+      "loss": 2.1278,
+      "step": 44000
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 0.0001011321316847907,
+      "loss": 2.118,
+      "step": 44500
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 9.889766125428273e-05,
+      "loss": 2.1219,
+      "step": 45000
+    },
+    {
+      "epoch": 6.7,
+      "eval_gen_len": 15.936,
+      "eval_loss": 2.5617337226867676,
+      "eval_rouge1": 29.8256,
+      "eval_rouge2": 11.3702,
+      "eval_rougeL": 26.4156,
+      "eval_rougeLsum": 26.8465,
+      "eval_runtime": 532.4007,
+      "eval_samples_per_second": 18.783,
+      "eval_steps_per_second": 2.348,
+      "step": 45000
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 9.666319082377476e-05,
+      "loss": 2.128,
+      "step": 45500
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 9.44287203932668e-05,
+      "loss": 2.1422,
+      "step": 46000
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 9.219424996275882e-05,
+      "loss": 2.1481,
+      "step": 46500
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 8.995977953225085e-05,
+      "loss": 2.1266,
+      "step": 47000
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 8.772530910174288e-05,
+      "loss": 2.0216,
+      "step": 47500
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 8.549083867123491e-05,
+      "loss": 2.0065,
+      "step": 48000
+    },
+    {
+      "epoch": 7.22,
+      "learning_rate": 8.325636824072694e-05,
+      "loss": 2.0364,
+      "step": 48500
+    },
+    {
+      "epoch": 7.3,
+      "learning_rate": 8.102189781021897e-05,
+      "loss": 2.0222,
+      "step": 49000
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 7.8787427379711e-05,
+      "loss": 2.0166,
+      "step": 49500
+    },
+    {
+      "epoch": 7.45,
+      "learning_rate": 7.655295694920303e-05,
+      "loss": 2.007,
+      "step": 50000
+    },
+    {
+      "epoch": 7.45,
+      "eval_gen_len": 15.7144,
+      "eval_loss": 2.6014492511749268,
+      "eval_rouge1": 29.743,
+      "eval_rouge2": 11.4336,
+      "eval_rougeL": 26.38,
+      "eval_rougeLsum": 26.772,
+      "eval_runtime": 532.2148,
+      "eval_samples_per_second": 18.789,
+      "eval_steps_per_second": 2.349,
+      "step": 50000
+    },
+    {
+      "epoch": 7.52,
+      "learning_rate": 7.431848651869506e-05,
+      "loss": 2.0252,
+      "step": 50500
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 7.20840160881871e-05,
+      "loss": 2.0376,
+      "step": 51000
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 6.984954565767912e-05,
+      "loss": 2.0143,
+      "step": 51500
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 6.761507522717115e-05,
+      "loss": 2.0242,
+      "step": 52000
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 6.538060479666319e-05,
+      "loss": 2.0094,
+      "step": 52500
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 6.314613436615521e-05,
+      "loss": 2.03,
+      "step": 53000
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 6.091166393564725e-05,
+      "loss": 2.02,
+      "step": 53500
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 5.867719350513928e-05,
+      "loss": 1.972,
+      "step": 54000
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 5.6442723074631304e-05,
+      "loss": 1.9319,
+      "step": 54500
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 5.4208252644123334e-05,
+      "loss": 1.9398,
+      "step": 55000
+    },
+    {
+      "epoch": 8.19,
+      "eval_gen_len": 15.9308,
+      "eval_loss": 2.608029365539551,
+      "eval_rouge1": 29.9478,
+      "eval_rouge2": 11.4801,
+      "eval_rougeL": 26.5352,
+      "eval_rougeLsum": 26.9746,
+      "eval_runtime": 532.8851,
+      "eval_samples_per_second": 18.766,
+      "eval_steps_per_second": 2.346,
+      "step": 55000
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 5.1973782213615364e-05,
+      "loss": 1.9379,
+      "step": 55500
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 4.9739311783107394e-05,
+      "loss": 1.9406,
+      "step": 56000
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 4.7504841352599425e-05,
+      "loss": 1.9487,
+      "step": 56500
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 4.5270370922091455e-05,
+      "loss": 1.9554,
+      "step": 57000
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 4.303590049158349e-05,
+      "loss": 1.9453,
+      "step": 57500
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 4.080143006107552e-05,
+      "loss": 1.9522,
+      "step": 58000
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 3.856695963056755e-05,
+      "loss": 1.958,
+      "step": 58500
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 3.633248920005958e-05,
+      "loss": 1.9495,
+      "step": 59000
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 3.409801876955161e-05,
+      "loss": 1.9276,
+      "step": 59500
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 3.186354833904364e-05,
+      "loss": 1.9426,
+      "step": 60000
+    },
+    {
+      "epoch": 8.94,
+      "eval_gen_len": 15.8598,
+      "eval_loss": 2.6021728515625,
+      "eval_rouge1": 30.097,
+      "eval_rouge2": 11.5602,
+      "eval_rougeL": 26.705,
+      "eval_rougeLsum": 27.1092,
+      "eval_runtime": 531.6176,
+      "eval_samples_per_second": 18.811,
+      "eval_steps_per_second": 2.351,
+      "step": 60000
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 2.9629077908535672e-05,
+      "loss": 1.9438,
+      "step": 60500
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 2.7394607478027705e-05,
+      "loss": 1.8807,
+      "step": 61000
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 2.5160137047519735e-05,
+      "loss": 1.8708,
+      "step": 61500
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 2.2925666617011765e-05,
+      "loss": 1.8727,
+      "step": 62000
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 2.0691196186503795e-05,
+      "loss": 1.8911,
+      "step": 62500
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 1.845672575599583e-05,
+      "loss": 1.8895,
+      "step": 63000
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 1.622225532548786e-05,
+      "loss": 1.8851,
+      "step": 63500
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 1.3987784894979889e-05,
+      "loss": 1.9082,
+      "step": 64000
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 1.1753314464471919e-05,
+      "loss": 1.8864,
+      "step": 64500
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 9.518844033963949e-06,
+      "loss": 1.8853,
+      "step": 65000
+    },
+    {
+      "epoch": 9.68,
+      "eval_gen_len": 15.803,
+      "eval_loss": 2.6137681007385254,
+      "eval_rouge1": 30.1588,
+      "eval_rouge2": 11.5823,
+      "eval_rougeL": 26.6984,
+      "eval_rougeLsum": 27.1371,
+      "eval_runtime": 529.3936,
+      "eval_samples_per_second": 18.89,
+      "eval_steps_per_second": 2.361,
+      "step": 65000
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 7.284373603455981e-06,
+      "loss": 1.8973,
+      "step": 65500
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 5.04990317294801e-06,
+      "loss": 1.8832,
+      "step": 66000
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 2.815432742440042e-06,
+      "loss": 1.8678,
+      "step": 66500
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 5.809623119320721e-07,
+      "loss": 1.9084,
+      "step": 67000
+    },
+    {
+      "epoch": 10.0,
+      "step": 67130,
+      "total_flos": 2.0764683296037274e+17,
+      "train_loss": 2.3876063485090873,
+      "train_runtime": 39696.1157,
+      "train_samples_per_second": 13.528,
+      "train_steps_per_second": 1.691
+    }
+  ],
+  "max_steps": 67130,
+  "num_train_epochs": 10,
+  "total_flos": 2.0764683296037274e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:954d8abdde4e1fcf58261d750d3e24dcdaabe264a9f1d1333d58551e441ddd3f
+size 3183