End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- id
 license: apache-2.0
 base_model: LazarusNLP/IndoNanoT5-base
 tags:
@@ -17,12 +19,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [LazarusNLP/IndoNanoT5-base](https://huggingface.co/LazarusNLP/IndoNanoT5-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5105
-- Rouge1: 72.6979
-- Rouge2: 65.6731
-- Rougel: 69.6184
-- Rougelsum: 71.7604
-- Gen Len: 102.2533
+- Loss: 0.5148
+- Rouge1: 72.9552
+- Rouge2: 66.0943
+- Rougel: 69.9638
+- Rougelsum: 72.0769
+- Gen Len: 102.96
 
 ## Model description
 

all_results.json

@@ -1,29 +1,29 @@
 {
     "epoch": 5.0,
-    "eval_gen_len": 101.78933333333333,
-    "eval_loss": 0.5441932678222656,
-    "eval_rouge1": 72.6224,
-    "eval_rouge2": 65.7072,
-    "eval_rougeL": 69.4293,
-    "eval_rougeLsum": 71.6325,
-    "eval_runtime": 988.0304,
+    "eval_gen_len": 102.96,
+    "eval_loss": 0.5148317217826843,
+    "eval_rouge1": 72.9552,
+    "eval_rouge2": 66.0943,
+    "eval_rougeL": 69.9638,
+    "eval_rougeLsum": 72.0769,
+    "eval_runtime": 985.0708,
     "eval_samples": 750,
-    "eval_samples_per_second": 0.759,
+    "eval_samples_per_second": 0.761,
     "eval_steps_per_second": 0.024,
-    "predict_gen_len": 101.4952025586354,
-    "predict_loss": 0.5291196703910828,
-    "predict_rouge1": 72.796,
-    "predict_rouge2": 66.0161,
-    "predict_rougeL": 69.9142,
-    "predict_rougeLsum": 71.9662,
-    "predict_runtime": 4936.0723,
+    "predict_gen_len": 102.63459488272922,
+    "predict_loss": 0.5010461211204529,
+    "predict_rouge1": 73.0775,
+    "predict_rouge2": 66.3491,
+    "predict_rougeL": 70.2443,
+    "predict_rougeLsum": 72.2686,
+    "predict_runtime": 4938.8836,
     "predict_samples": 3752,
     "predict_samples_per_second": 0.76,
     "predict_steps_per_second": 0.024,
-    "total_flos": 4.90582346563584e+16,
-    "train_loss": 0.6255005293362879,
-    "train_runtime": 4838.9085,
+    "total_flos": 4.925218464202752e+16,
+    "train_loss": 0.5920752076290113,
+    "train_runtime": 4819.5172,
     "train_samples": 14272,
-    "train_samples_per_second": 14.747,
-    "train_steps_per_second": 0.922
+    "train_samples_per_second": 14.806,
+    "train_steps_per_second": 0.925
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
-    "eval_gen_len": 101.78933333333333,
-    "eval_loss": 0.5441932678222656,
-    "eval_rouge1": 72.6224,
-    "eval_rouge2": 65.7072,
-    "eval_rougeL": 69.4293,
-    "eval_rougeLsum": 71.6325,
-    "eval_runtime": 988.0304,
+    "eval_gen_len": 102.96,
+    "eval_loss": 0.5148317217826843,
+    "eval_rouge1": 72.9552,
+    "eval_rouge2": 66.0943,
+    "eval_rougeL": 69.9638,
+    "eval_rougeLsum": 72.0769,
+    "eval_runtime": 985.0708,
     "eval_samples": 750,
-    "eval_samples_per_second": 0.759,
+    "eval_samples_per_second": 0.761,
     "eval_steps_per_second": 0.024
 }

predict_results.json

@@ -1,11 +1,11 @@
 {
-    "predict_gen_len": 101.4952025586354,
-    "predict_loss": 0.5291196703910828,
-    "predict_rouge1": 72.796,
-    "predict_rouge2": 66.0161,
-    "predict_rougeL": 69.9142,
-    "predict_rougeLsum": 71.9662,
-    "predict_runtime": 4936.0723,
+    "predict_gen_len": 102.63459488272922,
+    "predict_loss": 0.5010461211204529,
+    "predict_rouge1": 73.0775,
+    "predict_rouge2": 66.3491,
+    "predict_rougeL": 70.2443,
+    "predict_rougeLsum": 72.2686,
+    "predict_runtime": 4938.8836,
     "predict_samples": 3752,
     "predict_samples_per_second": 0.76,
     "predict_steps_per_second": 0.024

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 5.0,
-    "total_flos": 4.90582346563584e+16,
-    "train_loss": 0.6255005293362879,
-    "train_runtime": 4838.9085,
+    "total_flos": 4.925218464202752e+16,
+    "train_loss": 0.5920752076290113,
+    "train_runtime": 4819.5172,
     "train_samples": 14272,
-    "train_samples_per_second": 14.747,
-    "train_steps_per_second": 0.922
+    "train_samples_per_second": 14.806,
+    "train_steps_per_second": 0.925
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 72.3075,
-  "best_model_checkpoint": "bin/indosum-lora-3/checkpoint-2676",
+  "best_metric": 72.8512,
+  "best_model_checkpoint": "bin/indosum-lora-3/checkpoint-3568",
   "epoch": 5.0,
   "eval_steps": 500,
   "global_step": 4460,
@@ -10,112 +10,112 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.5876893401145935,
+      "grad_norm": 0.6515501141548157,
       "learning_rate": 0.0008,
-      "loss": 0.8274,
+      "loss": 0.7809,
       "step": 892
     },
     {
       "epoch": 1.0,
-      "eval_gen_len": 100.152,
-      "eval_loss": 0.5684335827827454,
-      "eval_rouge1": 69.5991,
-      "eval_rouge2": 62.0811,
-      "eval_rougeL": 66.2103,
-      "eval_rougeLsum": 68.5179,
-      "eval_runtime": 219.0061,
-      "eval_samples_per_second": 3.425,
+      "eval_gen_len": 100.42266666666667,
+      "eval_loss": 0.5510576963424683,
+      "eval_rouge1": 70.7387,
+      "eval_rouge2": 63.5003,
+      "eval_rougeL": 67.571,
+      "eval_rougeLsum": 69.7824,
+      "eval_runtime": 218.1822,
+      "eval_samples_per_second": 3.437,
       "eval_steps_per_second": 0.11,
       "step": 892
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.4317479431629181,
+      "grad_norm": 0.4482235908508301,
       "learning_rate": 0.0006,
-      "loss": 0.6237,
+      "loss": 0.5981,
       "step": 1784
     },
     {
       "epoch": 2.0,
-      "eval_gen_len": 104.78133333333334,
-      "eval_loss": 0.5486577153205872,
-      "eval_rouge1": 71.3379,
-      "eval_rouge2": 64.0461,
-      "eval_rougeL": 68.1399,
-      "eval_rougeLsum": 70.3424,
-      "eval_runtime": 219.5178,
-      "eval_samples_per_second": 3.417,
-      "eval_steps_per_second": 0.109,
+      "eval_gen_len": 104.784,
+      "eval_loss": 0.5352391600608826,
+      "eval_rouge1": 71.407,
+      "eval_rouge2": 64.2122,
+      "eval_rougeL": 68.3064,
+      "eval_rougeLsum": 70.4861,
+      "eval_runtime": 218.5017,
+      "eval_samples_per_second": 3.432,
+      "eval_steps_per_second": 0.11,
       "step": 1784
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.5349451303482056,
+      "grad_norm": 0.627132773399353,
       "learning_rate": 0.0004,
-      "loss": 0.5824,
+      "loss": 0.5542,
       "step": 2676
     },
     {
       "epoch": 3.0,
-      "eval_gen_len": 101.356,
-      "eval_loss": 0.5441932678222656,
-      "eval_rouge1": 72.3075,
-      "eval_rouge2": 65.3703,
-      "eval_rougeL": 69.1638,
-      "eval_rougeLsum": 71.3776,
-      "eval_runtime": 219.0561,
-      "eval_samples_per_second": 3.424,
-      "eval_steps_per_second": 0.11,
+      "eval_gen_len": 102.44266666666667,
+      "eval_loss": 0.5363225340843201,
+      "eval_rouge1": 72.3351,
+      "eval_rouge2": 65.2788,
+      "eval_rougeL": 69.2683,
+      "eval_rougeLsum": 71.4248,
+      "eval_runtime": 217.0288,
+      "eval_samples_per_second": 3.456,
+      "eval_steps_per_second": 0.111,
       "step": 2676
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.44907253980636597,
+      "grad_norm": 0.4163409471511841,
       "learning_rate": 0.0002,
-      "loss": 0.555,
+      "loss": 0.5238,
       "step": 3568
     },
     {
       "epoch": 4.0,
-      "eval_gen_len": 103.64666666666666,
-      "eval_loss": 0.5253363251686096,
-      "eval_rouge1": 72.2139,
-      "eval_rouge2": 65.2916,
-      "eval_rougeL": 69.2278,
-      "eval_rougeLsum": 71.2919,
-      "eval_runtime": 219.9659,
-      "eval_samples_per_second": 3.41,
-      "eval_steps_per_second": 0.109,
+      "eval_gen_len": 102.41733333333333,
+      "eval_loss": 0.5148317217826843,
+      "eval_rouge1": 72.8512,
+      "eval_rouge2": 65.9167,
+      "eval_rougeL": 69.8532,
+      "eval_rougeLsum": 71.9591,
+      "eval_runtime": 217.8801,
+      "eval_samples_per_second": 3.442,
+      "eval_steps_per_second": 0.11,
       "step": 3568
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.3915654420852661,
+      "grad_norm": 0.39186742901802063,
       "learning_rate": 0.0,
-      "loss": 0.5391,
+      "loss": 0.5034,
       "step": 4460
     },
     {
       "epoch": 5.0,
-      "eval_gen_len": 102.11733333333333,
-      "eval_loss": 0.5204364657402039,
-      "eval_rouge1": 72.0194,
-      "eval_rouge2": 64.9755,
-      "eval_rougeL": 68.9244,
-      "eval_rougeLsum": 71.0549,
-      "eval_runtime": 218.95,
-      "eval_samples_per_second": 3.425,
+      "eval_gen_len": 102.25333333333333,
+      "eval_loss": 0.5104866027832031,
+      "eval_rouge1": 72.6979,
+      "eval_rouge2": 65.6731,
+      "eval_rougeL": 69.6184,
+      "eval_rougeLsum": 71.7604,
+      "eval_runtime": 218.1186,
+      "eval_samples_per_second": 3.438,
       "eval_steps_per_second": 0.11,
       "step": 4460
     },
     {
       "epoch": 5.0,
       "step": 4460,
-      "total_flos": 4.90582346563584e+16,
-      "train_loss": 0.6255005293362879,
-      "train_runtime": 4838.9085,
-      "train_samples_per_second": 14.747,
-      "train_steps_per_second": 0.922
+      "total_flos": 4.925218464202752e+16,
+      "train_loss": 0.5920752076290113,
+      "train_runtime": 4819.5172,
+      "train_samples_per_second": 14.806,
+      "train_steps_per_second": 0.925
     }
   ],
   "logging_steps": 500,
@@ -123,7 +123,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 4.90582346563584e+16,
+  "total_flos": 4.925218464202752e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null