End of training

README.md

@@ -1,11 +1,24 @@
 ---
+base_model: /exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210
 tags:
 - generated_from_trainer
+datasets:
+- learn3r/summ_screen_fd_bp
 metrics:
 - rouge
 model-index:
 - name: longt5_xl_summ_screen_bp_only_30
-  results: []
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: learn3r/summ_screen_fd_bp
+      type: learn3r/summ_screen_fd_bp
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 40.4388
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -13,14 +26,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # longt5_xl_summ_screen_bp_only_30
 
-This model was trained from scratch on an unknown dataset.
+This model is a fine-tuned version of [/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210](https://huggingface.co//exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210) on the learn3r/summ_screen_fd_bp dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.4990
-- Rouge1: 32.5815
-- Rouge2: 14.2951
-- Rougel: 22.4501
-- Rougelsum: 31.2928
-- Gen Len: 499.3107
+- Loss: 2.2376
+- Rouge1: 40.4388
+- Rouge2: 16.4662
+- Rougel: 28.0771
+- Rougelsum: 38.3405
+- Gen Len: 246.7396
 
 ## Model description
 

all_results.json

@@ -1,18 +1,18 @@
 {
-    "epoch": 13.18,
-    "eval_gen_len": 246.3491124260355,
-    "eval_loss": 2.2397100925445557,
-    "eval_rouge1": 40.4943,
-    "eval_rouge2": 16.4695,
-    "eval_rougeL": 28.0964,
-    "eval_rougeLsum": 38.3693,
-    "eval_runtime": 467.1505,
+    "epoch": 14.61,
+    "eval_gen_len": 246.7396449704142,
+    "eval_loss": 2.237640380859375,
+    "eval_rouge1": 40.4388,
+    "eval_rouge2": 16.4662,
+    "eval_rougeL": 28.0771,
+    "eval_rougeLsum": 38.3405,
+    "eval_runtime": 1708.4379,
     "eval_samples": 338,
-    "eval_samples_per_second": 0.724,
-    "eval_steps_per_second": 0.024,
-    "train_loss": 0.16495737830797833,
-    "train_runtime": 12728.0808,
+    "eval_samples_per_second": 0.198,
+    "eval_steps_per_second": 0.025,
+    "train_loss": 0.13047422234501158,
+    "train_runtime": 78176.4332,
     "train_samples": 3673,
-    "train_samples_per_second": 4.329,
-    "train_steps_per_second": 0.004
+    "train_samples_per_second": 0.705,
+    "train_steps_per_second": 0.003
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 13.18,
-    "eval_gen_len": 246.3491124260355,
-    "eval_loss": 2.2397100925445557,
-    "eval_rouge1": 40.4943,
-    "eval_rouge2": 16.4695,
-    "eval_rougeL": 28.0964,
-    "eval_rougeLsum": 38.3693,
-    "eval_runtime": 467.1505,
+    "epoch": 14.61,
+    "eval_gen_len": 246.7396449704142,
+    "eval_loss": 2.237640380859375,
+    "eval_rouge1": 40.4388,
+    "eval_rouge2": 16.4662,
+    "eval_rougeL": 28.0771,
+    "eval_rougeLsum": 38.3405,
+    "eval_runtime": 1708.4379,
     "eval_samples": 338,
-    "eval_samples_per_second": 0.724,
-    "eval_steps_per_second": 0.024
+    "eval_samples_per_second": 0.198,
+    "eval_steps_per_second": 0.025
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 13.18,
-    "train_loss": 0.16495737830797833,
-    "train_runtime": 12728.0808,
+    "epoch": 14.61,
+    "train_loss": 0.13047422234501158,
+    "train_runtime": 78176.4332,
     "train_samples": 3673,
-    "train_samples_per_second": 4.329,
-    "train_steps_per_second": 0.004
+    "train_samples_per_second": 0.705,
+    "train_steps_per_second": 0.003
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 2.237640380859375,
   "best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only_30/checkpoint-14",
-  "epoch": 13.182608695652174,
+  "epoch": 14.608695652173914,
   "eval_steps": 500,
-  "global_step": 45,
+  "global_step": 210,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -58,246 +58,796 @@
       "eval_rouge2": 16.4662,
       "eval_rougeL": 28.0771,
       "eval_rougeLsum": 38.3405,
-      "eval_runtime": 1741.7742,
-      "eval_samples_per_second": 0.194,
+      "eval_runtime": 1709.2671,
+      "eval_samples_per_second": 0.198,
       "eval_steps_per_second": 0.025,
       "step": 14
     },
     {
-      "epoch": 4.83,
-      "eval_gen_len": 503.56508875739644,
-      "eval_loss": 2.572709798812866,
-      "eval_rouge1": 30.0123,
-      "eval_rouge2": 12.3701,
-      "eval_rougeL": 21.2834,
-      "eval_rougeLsum": 28.891,
-      "eval_runtime": 472.1582,
-      "eval_samples_per_second": 0.716,
-      "eval_steps_per_second": 0.023,
-      "step": 15
-    },
-    {
-      "epoch": 5.11,
+      "epoch": 1.11,
       "learning_rate": 0.0005,
-      "loss": 0.278,
+      "loss": 0.2825,
       "step": 16
     },
     {
-      "epoch": 5.67,
+      "epoch": 1.25,
       "learning_rate": 0.0005,
-      "loss": 0.3036,
+      "loss": 0.2835,
       "step": 18
     },
     {
-      "epoch": 5.95,
-      "eval_gen_len": 506.9526627218935,
-      "eval_loss": 2.2658941745758057,
-      "eval_rouge1": 27.8421,
-      "eval_rouge2": 11.1942,
-      "eval_rougeL": 20.4713,
-      "eval_rougeLsum": 26.6097,
-      "eval_runtime": 472.9998,
-      "eval_samples_per_second": 0.715,
-      "eval_steps_per_second": 0.023,
-      "step": 19
-    },
-    {
-      "epoch": 6.23,
+      "epoch": 1.39,
       "learning_rate": 0.0005,
-      "loss": 0.2937,
+      "loss": 0.2732,
       "step": 20
     },
     {
-      "epoch": 6.78,
+      "epoch": 1.53,
       "learning_rate": 0.0005,
-      "loss": 0.2941,
+      "loss": 0.2928,
       "step": 22
     },
     {
-      "epoch": 6.78,
-      "eval_gen_len": 284.2307692307692,
-      "eval_loss": 2.2636122703552246,
-      "eval_rouge1": 40.8304,
-      "eval_rouge2": 17.3615,
-      "eval_rougeL": 28.0971,
-      "eval_rougeLsum": 39.0943,
-      "eval_runtime": 468.1667,
-      "eval_samples_per_second": 0.722,
-      "eval_steps_per_second": 0.023,
-      "step": 22
-    },
-    {
-      "epoch": 7.34,
+      "epoch": 1.67,
       "learning_rate": 0.0005,
-      "loss": 0.2508,
+      "loss": 0.2912,
       "step": 24
     },
     {
-      "epoch": 7.9,
+      "epoch": 1.81,
       "learning_rate": 0.0005,
-      "loss": 0.2642,
-      "step": 26
-    },
-    {
-      "epoch": 7.9,
-      "eval_gen_len": 341.25147928994085,
-      "eval_loss": 2.286357879638672,
-      "eval_rouge1": 38.3377,
-      "eval_rouge2": 15.8119,
-      "eval_rougeL": 26.4838,
-      "eval_rougeLsum": 36.5174,
-      "eval_runtime": 469.1516,
-      "eval_samples_per_second": 0.72,
-      "eval_steps_per_second": 0.023,
+      "loss": 0.2914,
       "step": 26
     },
     {
-      "epoch": 8.45,
+      "epoch": 1.95,
       "learning_rate": 0.0005,
-      "loss": 0.2604,
+      "loss": 0.2707,
       "step": 28
     },
     {
-      "epoch": 8.73,
-      "eval_gen_len": 435.26331360946745,
-      "eval_loss": 2.455064535140991,
-      "eval_rouge1": 33.2021,
-      "eval_rouge2": 13.6577,
-      "eval_rougeL": 23.3288,
-      "eval_rougeLsum": 31.8326,
-      "eval_runtime": 471.959,
-      "eval_samples_per_second": 0.716,
-      "eval_steps_per_second": 0.023,
-      "step": 29
+      "epoch": 1.95,
+      "eval_gen_len": 307.3786982248521,
+      "eval_loss": 2.320437431335449,
+      "eval_rouge1": 40.2873,
+      "eval_rouge2": 16.7641,
+      "eval_rougeL": 27.3895,
+      "eval_rougeLsum": 38.2689,
+      "eval_runtime": 1775.692,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 28
     },
     {
-      "epoch": 9.01,
+      "epoch": 2.09,
       "learning_rate": 0.0005,
-      "loss": 0.2422,
+      "loss": 0.2342,
       "step": 30
     },
     {
-      "epoch": 9.57,
+      "epoch": 2.23,
       "learning_rate": 0.0005,
-      "loss": 0.2237,
+      "loss": 0.2326,
       "step": 32
     },
     {
-      "epoch": 9.84,
-      "eval_gen_len": 234.61242603550295,
-      "eval_loss": 2.615253210067749,
-      "eval_rouge1": 40.3297,
-      "eval_rouge2": 15.3786,
-      "eval_rougeL": 28.1208,
-      "eval_rougeLsum": 38.2426,
-      "eval_runtime": 470.1581,
-      "eval_samples_per_second": 0.719,
-      "eval_steps_per_second": 0.023,
-      "step": 33
-    },
-    {
-      "epoch": 10.12,
+      "epoch": 2.37,
       "learning_rate": 0.0005,
-      "loss": 0.2184,
+      "loss": 0.2334,
       "step": 34
     },
     {
-      "epoch": 10.68,
+      "epoch": 2.5,
       "learning_rate": 0.0005,
-      "loss": 0.1904,
+      "loss": 0.2212,
       "step": 36
     },
     {
-      "epoch": 10.96,
-      "eval_gen_len": 174.57396449704143,
-      "eval_loss": 2.666517734527588,
-      "eval_rouge1": 39.6006,
-      "eval_rouge2": 14.9586,
-      "eval_rougeL": 27.2453,
-      "eval_rougeLsum": 37.6744,
-      "eval_runtime": 450.9467,
-      "eval_samples_per_second": 0.75,
+      "epoch": 2.64,
+      "learning_rate": 0.0005,
+      "loss": 0.2161,
+      "step": 38
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.0005,
+      "loss": 0.2236,
+      "step": 40
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.0005,
+      "loss": 0.2217,
+      "step": 42
+    },
+    {
+      "epoch": 2.99,
+      "eval_gen_len": 501.93195266272187,
+      "eval_loss": 2.5281190872192383,
+      "eval_rouge1": 31.9916,
+      "eval_rouge2": 13.8136,
+      "eval_rougeL": 22.1895,
+      "eval_rougeLsum": 30.623,
+      "eval_runtime": 1780.5368,
+      "eval_samples_per_second": 0.19,
       "eval_steps_per_second": 0.024,
-      "step": 37
+      "step": 43
     },
     {
-      "epoch": 11.23,
+      "epoch": 3.06,
       "learning_rate": 0.0005,
-      "loss": 0.2076,
-      "step": 38
+      "loss": 0.2055,
+      "step": 44
     },
     {
-      "epoch": 11.79,
+      "epoch": 3.2,
       "learning_rate": 0.0005,
-      "loss": 0.2247,
-      "step": 40
+      "loss": 0.1943,
+      "step": 46
     },
     {
-      "epoch": 11.79,
-      "eval_gen_len": 500.58284023668637,
-      "eval_loss": 2.722362518310547,
-      "eval_rouge1": 30.5957,
-      "eval_rouge2": 13.3496,
-      "eval_rougeL": 21.9712,
-      "eval_rougeLsum": 29.22,
-      "eval_runtime": 477.6586,
-      "eval_samples_per_second": 0.708,
-      "eval_steps_per_second": 0.023,
-      "step": 40
+      "epoch": 3.34,
+      "learning_rate": 0.0005,
+      "loss": 0.1875,
+      "step": 48
     },
     {
-      "epoch": 12.35,
+      "epoch": 3.48,
       "learning_rate": 0.0005,
-      "loss": 0.1866,
-      "step": 42
+      "loss": 0.1909,
+      "step": 50
     },
     {
-      "epoch": 12.9,
+      "epoch": 3.62,
       "learning_rate": 0.0005,
-      "loss": 0.182,
-      "step": 44
+      "loss": 0.1881,
+      "step": 52
     },
     {
-      "epoch": 12.9,
-      "eval_gen_len": 259.6568047337278,
-      "eval_loss": 3.271503448486328,
-      "eval_rouge1": 41.6828,
-      "eval_rouge2": 17.0818,
-      "eval_rougeL": 28.087,
-      "eval_rougeLsum": 39.5947,
-      "eval_runtime": 468.2637,
-      "eval_samples_per_second": 0.722,
-      "eval_steps_per_second": 0.023,
-      "step": 44
+      "epoch": 3.76,
+      "learning_rate": 0.0005,
+      "loss": 0.1541,
+      "step": 54
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 0.0005,
+      "loss": 0.1776,
+      "step": 56
+    },
+    {
+      "epoch": 3.97,
+      "eval_gen_len": 489.6183431952663,
+      "eval_loss": 2.7530014514923096,
+      "eval_rouge1": 31.7535,
+      "eval_rouge2": 13.8852,
+      "eval_rougeL": 22.8653,
+      "eval_rougeLsum": 30.3796,
+      "eval_runtime": 1780.5033,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 57
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 0.0005,
+      "loss": 0.1876,
+      "step": 58
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 0.0005,
+      "loss": 0.1589,
+      "step": 60
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 0.0005,
+      "loss": 0.1529,
+      "step": 62
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 0.0005,
+      "loss": 0.1464,
+      "step": 64
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.0005,
+      "loss": 0.1689,
+      "step": 66
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 0.0005,
+      "loss": 0.1492,
+      "step": 68
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 0.0005,
+      "loss": 0.1424,
+      "step": 70
+    },
+    {
+      "epoch": 4.94,
+      "eval_gen_len": 502.11242603550295,
+      "eval_loss": 2.657783031463623,
+      "eval_rouge1": 32.117,
+      "eval_rouge2": 14.2141,
+      "eval_rougeL": 22.3733,
+      "eval_rougeLsum": 30.8328,
+      "eval_runtime": 1780.0926,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 71
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 0.0005,
+      "loss": 0.1494,
+      "step": 72
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 0.0005,
+      "loss": 0.118,
+      "step": 74
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 0.0005,
+      "loss": 0.1499,
+      "step": 76
+    },
+    {
+      "epoch": 5.43,
+      "learning_rate": 0.0005,
+      "loss": 0.1369,
+      "step": 78
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 0.0005,
+      "loss": 0.1084,
+      "step": 80
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 0.0005,
+      "loss": 0.117,
+      "step": 82
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 0.0005,
+      "loss": 0.144,
+      "step": 84
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 0.0005,
+      "loss": 0.1449,
+      "step": 86
+    },
+    {
+      "epoch": 5.98,
+      "eval_gen_len": 357.31360946745565,
+      "eval_loss": 2.5508346557617188,
+      "eval_rouge1": 35.3448,
+      "eval_rouge2": 13.8478,
+      "eval_rougeL": 24.9044,
+      "eval_rougeLsum": 33.6108,
+      "eval_runtime": 1768.966,
+      "eval_samples_per_second": 0.191,
+      "eval_steps_per_second": 0.024,
+      "step": 86
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 0.0005,
+      "loss": 0.1101,
+      "step": 88
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 0.0005,
+      "loss": 0.0985,
+      "step": 90
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 0.0005,
+      "loss": 0.1101,
+      "step": 92
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 0.0005,
+      "loss": 0.1013,
+      "step": 94
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 0.0005,
+      "loss": 0.1057,
+      "step": 96
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 0.0005,
+      "loss": 0.1102,
+      "step": 98
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 0.0005,
+      "loss": 0.1191,
+      "step": 100
+    },
+    {
+      "epoch": 6.96,
+      "eval_gen_len": 408.86686390532543,
+      "eval_loss": 3.162177324295044,
+      "eval_rouge1": 37.2189,
+      "eval_rouge2": 16.0076,
+      "eval_rougeL": 25.7011,
+      "eval_rougeLsum": 35.294,
+      "eval_runtime": 1778.8704,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 100
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 0.0005,
+      "loss": 0.1,
+      "step": 102
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 0.0005,
+      "loss": 0.0949,
+      "step": 104
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 0.0005,
+      "loss": 0.0974,
+      "step": 106
+    },
+    {
+      "epoch": 7.51,
+      "learning_rate": 0.0005,
+      "loss": 0.096,
+      "step": 108
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 0.0005,
+      "loss": 0.1023,
+      "step": 110
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 0.0005,
+      "loss": 0.0829,
+      "step": 112
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 0.0005,
+      "loss": 0.0879,
+      "step": 114
+    },
+    {
+      "epoch": 8.0,
+      "eval_gen_len": 318.2278106508876,
+      "eval_loss": 2.8510310649871826,
+      "eval_rouge1": 39.8825,
+      "eval_rouge2": 16.8073,
+      "eval_rougeL": 27.2428,
+      "eval_rougeLsum": 37.9568,
+      "eval_runtime": 1776.4341,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 115
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 0.0005,
+      "loss": 0.0957,
+      "step": 116
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 0.0005,
+      "loss": 0.0991,
+      "step": 118
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 0.0005,
+      "loss": 0.095,
+      "step": 120
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 0.0005,
+      "loss": 0.0999,
+      "step": 122
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 0.0005,
+      "loss": 0.0787,
+      "step": 124
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 0.0005,
+      "loss": 0.0716,
+      "step": 126
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.0005,
+      "loss": 0.0899,
+      "step": 128
+    },
+    {
+      "epoch": 8.97,
+      "eval_gen_len": 500.405325443787,
+      "eval_loss": 2.9137675762176514,
+      "eval_rouge1": 31.7139,
+      "eval_rouge2": 13.7066,
+      "eval_rougeL": 21.8844,
+      "eval_rougeLsum": 30.5075,
+      "eval_runtime": 1780.3113,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 129
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.0005,
+      "loss": 0.0945,
+      "step": 130
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 0.0005,
+      "loss": 0.0597,
+      "step": 132
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 0.0005,
+      "loss": 0.0586,
+      "step": 134
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 0.0005,
+      "loss": 0.0757,
+      "step": 136
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 0.0005,
+      "loss": 0.0665,
+      "step": 138
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 0.0005,
+      "loss": 0.0594,
+      "step": 140
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 0.0005,
+      "loss": 0.0656,
+      "step": 142
+    },
+    {
+      "epoch": 9.95,
+      "eval_gen_len": 488.1686390532544,
+      "eval_loss": 3.1616334915161133,
+      "eval_rouge1": 33.055,
+      "eval_rouge2": 14.5841,
+      "eval_rougeL": 22.5883,
+      "eval_rougeLsum": 31.7565,
+      "eval_runtime": 1782.2406,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 143
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 0.0005,
+      "loss": 0.0613,
+      "step": 144
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 0.0005,
+      "loss": 0.0621,
+      "step": 146
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 0.0005,
+      "loss": 0.0739,
+      "step": 148
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 0.0005,
+      "loss": 0.0841,
+      "step": 150
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 0.0005,
+      "loss": 0.0562,
+      "step": 152
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 0.0005,
+      "loss": 0.0611,
+      "step": 154
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 0.0005,
+      "loss": 0.0607,
+      "step": 156
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 0.0005,
+      "loss": 0.0542,
+      "step": 158
+    },
+    {
+      "epoch": 10.99,
+      "eval_gen_len": 198.80769230769232,
+      "eval_loss": 3.3630056381225586,
+      "eval_rouge1": 43.7514,
+      "eval_rouge2": 18.9011,
+      "eval_rougeL": 29.9017,
+      "eval_rougeLsum": 41.6887,
+      "eval_runtime": 1460.3937,
+      "eval_samples_per_second": 0.231,
+      "eval_steps_per_second": 0.029,
+      "step": 158
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 0.0005,
+      "loss": 0.0548,
+      "step": 160
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 0.0005,
+      "loss": 0.0477,
+      "step": 162
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 0.0005,
+      "loss": 0.052,
+      "step": 164
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 0.0005,
+      "loss": 0.053,
+      "step": 166
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 0.0005,
+      "loss": 0.0525,
+      "step": 168
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 0.0005,
+      "loss": 0.0555,
+      "step": 170
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 0.0005,
+      "loss": 0.0557,
+      "step": 172
+    },
+    {
+      "epoch": 11.97,
+      "eval_gen_len": 270.9674556213018,
+      "eval_loss": 3.3825831413269043,
+      "eval_rouge1": 42.3089,
+      "eval_rouge2": 18.2735,
+      "eval_rougeL": 29.0356,
+      "eval_rougeLsum": 40.4154,
+      "eval_runtime": 1704.6973,
+      "eval_samples_per_second": 0.198,
+      "eval_steps_per_second": 0.025,
+      "step": 172
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 0.0005,
+      "loss": 0.0513,
+      "step": 174
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 0.0005,
+      "loss": 0.0419,
+      "step": 176
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 0.0005,
+      "loss": 0.0525,
+      "step": 178
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 0.0005,
+      "loss": 0.0535,
+      "step": 180
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 0.0005,
+      "loss": 0.0611,
+      "step": 182
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 0.0005,
+      "loss": 0.0444,
+      "step": 184
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 0.0005,
+      "loss": 0.0542,
+      "step": 186
+    },
+    {
+      "epoch": 12.94,
+      "eval_gen_len": 186.73076923076923,
+      "eval_loss": 3.4408490657806396,
+      "eval_rouge1": 40.7691,
+      "eval_rouge2": 16.529,
+      "eval_rougeL": 28.3999,
+      "eval_rougeLsum": 38.9723,
+      "eval_runtime": 1525.6668,
+      "eval_samples_per_second": 0.222,
+      "eval_steps_per_second": 0.028,
+      "step": 186
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 0.0005,
+      "loss": 0.0602,
+      "step": 188
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 0.0005,
+      "loss": 0.0438,
+      "step": 190
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 0.0005,
+      "loss": 0.0503,
+      "step": 192
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 0.0005,
+      "loss": 0.046,
+      "step": 194
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 0.0005,
+      "loss": 0.0368,
+      "step": 196
+    },
+    {
+      "epoch": 13.77,
+      "learning_rate": 0.0005,
+      "loss": 0.0572,
+      "step": 198
+    },
+    {
+      "epoch": 13.91,
+      "learning_rate": 0.0005,
+      "loss": 0.0596,
+      "step": 200
+    },
+    {
+      "epoch": 13.98,
+      "eval_gen_len": 398.4704142011834,
+      "eval_loss": 3.525272846221924,
+      "eval_rouge1": 37.0037,
+      "eval_rouge2": 15.9098,
+      "eval_rougeL": 25.2808,
+      "eval_rougeLsum": 35.3868,
+      "eval_runtime": 1778.3289,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 201
+    },
+    {
+      "epoch": 14.05,
+      "learning_rate": 0.0005,
+      "loss": 0.0434,
+      "step": 202
+    },
+    {
+      "epoch": 14.19,
+      "learning_rate": 0.0005,
+      "loss": 0.0453,
+      "step": 204
+    },
+    {
+      "epoch": 14.33,
+      "learning_rate": 0.0005,
+      "loss": 0.0453,
+      "step": 206
+    },
+    {
+      "epoch": 14.47,
+      "learning_rate": 0.0005,
+      "loss": 0.0586,
+      "step": 208
+    },
+    {
+      "epoch": 14.61,
+      "learning_rate": 0.0005,
+      "loss": 0.0385,
+      "step": 210
+    },
+    {
+      "epoch": 14.61,
+      "eval_gen_len": 499.31065088757396,
+      "eval_loss": 3.498972177505493,
+      "eval_rouge1": 32.5815,
+      "eval_rouge2": 14.2951,
+      "eval_rougeL": 22.4501,
+      "eval_rougeLsum": 31.2928,
+      "eval_runtime": 1779.9602,
+      "eval_samples_per_second": 0.19,
+      "eval_steps_per_second": 0.024,
+      "step": 210
     },
     {
-      "epoch": 13.18,
-      "eval_gen_len": 484.396449704142,
-      "eval_loss": 2.397312641143799,
-      "eval_rouge1": 31.9833,
-      "eval_rouge2": 14.0141,
-      "eval_rougeL": 22.6823,
-      "eval_rougeLsum": 30.6424,
-      "eval_runtime": 471.2439,
-      "eval_samples_per_second": 0.717,
-      "eval_steps_per_second": 0.023,
-      "step": 45
-    },
-    {
-      "epoch": 13.18,
-      "step": 45,
-      "total_flos": 2.4232217682745754e+18,
-      "train_loss": 0.16495737830797833,
-      "train_runtime": 12728.0808,
-      "train_samples_per_second": 4.329,
-      "train_steps_per_second": 0.004
+      "epoch": 14.61,
+      "step": 210,
+      "total_flos": 3.6715371967648973e+18,
+      "train_loss": 0.13047422234501158,
+      "train_runtime": 78176.4332,
+      "train_samples_per_second": 0.705,
+      "train_steps_per_second": 0.003
     }
   ],
   "logging_steps": 2,
-  "max_steps": 45,
+  "max_steps": 210,
   "num_train_epochs": 15,
   "save_steps": 500,
-  "total_flos": 2.4232217682745754e+18,
+  "total_flos": 3.6715371967648973e+18,
   "trial_name": null,
   "trial_params": null
 }