Training in progress, epoch 1

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 3.0,
-    "total_flos": 3.9828842929402675e+17,
-    "train_loss": 1.6144965134510645,
-    "train_runtime": 36481.981,
-    "train_samples": 193883,
-    "train_samples_per_second": 15.943,
-    "train_steps_per_second": 0.996
+    "total_flos": 2054272057344000.0,
+    "train_loss": 2.2623963128952753,
+    "train_runtime": 857.0722,
+    "train_samples": 1000,
+    "train_samples_per_second": 3.5,
+    "train_steps_per_second": 0.221
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:160c9799444b23dd770020cbe9c87bb351b6c08a13e80411dee39ee5195d2713
+oid sha256:7311009544072c5a7bce8c227252e2c97825e1c6c54482e043d0ce4210b04909
 size 990348536

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 3.0,
-    "total_flos": 3.9828842929402675e+17,
-    "train_loss": 1.6144965134510645,
-    "train_runtime": 36481.981,
-    "train_samples": 193883,
-    "train_samples_per_second": 15.943,
-    "train_steps_per_second": 0.996
+    "total_flos": 2054272057344000.0,
+    "train_loss": 2.2623963128952753,
+    "train_runtime": 857.0722,
+    "train_samples": 1000,
+    "train_samples_per_second": 3.5,
+    "train_steps_per_second": 0.221
 }

trainer_state.json

@@ -1,89 +1,89 @@
 {
-  "best_metric": 35.4918,
-  "best_model_checkpoint": "bin/liputan6-base/checkpoint-36354",
+  "best_metric": 19.886,
+  "best_model_checkpoint": "bin/liputan6-base/checkpoint-189",
   "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 36354,
+  "global_step": 189,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.8706055879592896,
+      "grad_norm": 5.871487617492676,
       "learning_rate": 0.0006666666666666666,
-      "loss": 2.0683,
-      "step": 12118
+      "loss": 3.7824,
+      "step": 63
     },
     {
       "epoch": 1.0,
-      "eval_gen_len": 38.11319722931098,
-      "eval_loss": 2.372281312942505,
-      "eval_rouge1": 33.423,
-      "eval_rouge2": 16.5875,
-      "eval_rougeL": 28.0723,
-      "eval_rougeLsum": 31.0371,
-      "eval_runtime": 1335.7227,
-      "eval_samples_per_second": 8.214,
-      "eval_steps_per_second": 0.129,
-      "step": 12118
+      "eval_gen_len": 31.998,
+      "eval_loss": 3.822542190551758,
+      "eval_rouge1": 16.8681,
+      "eval_rouge2": 5.1649,
+      "eval_rougeL": 14.9393,
+      "eval_rougeLsum": 15.6881,
+      "eval_runtime": 235.6514,
+      "eval_samples_per_second": 4.244,
+      "eval_steps_per_second": 0.136,
+      "step": 63
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.7888175845146179,
+      "grad_norm": 4.992724895477295,
       "learning_rate": 0.0003333333333333333,
-      "loss": 1.5342,
-      "step": 24236
+      "loss": 2.0598,
+      "step": 126
     },
     {
       "epoch": 2.0,
-      "eval_gen_len": 36.71764491432738,
-      "eval_loss": 2.139983654022217,
-      "eval_rouge1": 34.4775,
-      "eval_rouge2": 17.1757,
-      "eval_rougeL": 28.8546,
-      "eval_rougeLsum": 32.0264,
-      "eval_runtime": 1116.2631,
-      "eval_samples_per_second": 9.829,
-      "eval_steps_per_second": 0.154,
-      "step": 24236
+      "eval_gen_len": 38.822,
+      "eval_loss": 3.8557004928588867,
+      "eval_rouge1": 18.6028,
+      "eval_rouge2": 5.9301,
+      "eval_rougeL": 15.9454,
+      "eval_rougeLsum": 17.1143,
+      "eval_runtime": 247.4796,
+      "eval_samples_per_second": 4.041,
+      "eval_steps_per_second": 0.129,
+      "step": 126
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.6657369136810303,
+      "grad_norm": 4.042899131774902,
       "learning_rate": 0.0,
-      "loss": 1.241,
-      "step": 36354
+      "loss": 0.945,
+      "step": 189
     },
     {
       "epoch": 3.0,
-      "eval_gen_len": 36.4740247903755,
-      "eval_loss": 2.039839029312134,
-      "eval_rouge1": 35.4918,
-      "eval_rouge2": 17.885,
-      "eval_rougeL": 29.6804,
-      "eval_rougeLsum": 32.9507,
-      "eval_runtime": 1054.5787,
-      "eval_samples_per_second": 10.404,
-      "eval_steps_per_second": 0.163,
-      "step": 36354
+      "eval_gen_len": 30.875,
+      "eval_loss": 4.205010414123535,
+      "eval_rouge1": 19.886,
+      "eval_rouge2": 6.0542,
+      "eval_rougeL": 16.8703,
+      "eval_rougeLsum": 18.2573,
+      "eval_runtime": 188.2411,
+      "eval_samples_per_second": 5.312,
+      "eval_steps_per_second": 0.17,
+      "step": 189
     },
     {
       "epoch": 3.0,
-      "step": 36354,
-      "total_flos": 3.9828842929402675e+17,
-      "train_loss": 1.6144965134510645,
-      "train_runtime": 36481.981,
-      "train_samples_per_second": 15.943,
-      "train_steps_per_second": 0.996
+      "step": 189,
+      "total_flos": 2054272057344000.0,
+      "train_loss": 2.2623963128952753,
+      "train_runtime": 857.0722,
+      "train_samples_per_second": 3.5,
+      "train_steps_per_second": 0.221
     }
   ],
   "logging_steps": 500,
-  "max_steps": 36354,
+  "max_steps": 189,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 3.9828842929402675e+17,
+  "total_flos": 2054272057344000.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4caf5727664b0a9127d4155c25dd62a1c519c7a98850d447fb40354f555cd16
+oid sha256:3d29847bbc10a63ad9d7df3718582986264be90f86beb188d38e2c6cb7b637d9
 size 5176