Training in progress, epoch 1, checkpoint

checkpoint-1528/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e046d25e4ad5c74b91be4fd3536533fc0663fa47e08b72382d149f3176a5d205
+oid sha256:74e357ce0b7042a6f40e5f45c53ad211d929476c8e3445c880e62339e20b708b
 size 557116312

checkpoint-1528/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae7c16a498d582599cd05fbab821d1100a28ed8fbbbe5e414bac30bd70fe8299
+oid sha256:0fe26fcb5c0e353bf67a52789fb4cfae7be61e381b062d125ca91b1c604de9b5
 size 1113991930

checkpoint-1528/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8e6dccea0da7d7eff4bfe6269cc5b94b43442b566ea0f37db1b49e1824d969a
+oid sha256:2a1c88126ef67a930b2cd0c54619e80f9ef80c6abae00cbf436ab3946bd2d2c0
 size 1064

checkpoint-1528/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 25.4493,
-  "best_model_checkpoint": "my-model/checkpoint-1528",
+  "best_metric": 25.2705,
+  "best_model_checkpoint": "./ArabartModel-1/checkpoint-1528",
   "epoch": 1.0,
   "eval_steps": 500,
   "global_step": 1528,
@@ -10,41 +10,41 @@
   "log_history": [
     {
       "epoch": 0.32722513089005234,
-      "grad_norm": 0.41872361302375793,
-      "learning_rate": 4.836387434554974e-05,
-      "loss": 0.0155,
+      "grad_norm": 1.563672423362732,
+      "learning_rate": 4.6727748691099475e-05,
+      "loss": 0.0195,
       "step": 500
     },
     {
       "epoch": 0.6544502617801047,
-      "grad_norm": 2.3573551177978516,
-      "learning_rate": 4.6727748691099475e-05,
-      "loss": 0.0193,
+      "grad_norm": 1.7397737503051758,
+      "learning_rate": 4.3455497382198955e-05,
+      "loss": 0.0244,
       "step": 1000
     },
     {
       "epoch": 0.981675392670157,
-      "grad_norm": 2.1439473628997803,
-      "learning_rate": 4.5091623036649215e-05,
-      "loss": 0.0231,
+      "grad_norm": 3.006633758544922,
+      "learning_rate": 4.018324607329843e-05,
+      "loss": 0.0293,
       "step": 1500
     },
     {
       "epoch": 1.0,
-      "eval_loss": 5.686821937561035,
-      "eval_rouge1": 25.4493,
-      "eval_rouge2": 10.1381,
-      "eval_rougeL": 22.2977,
-      "eval_runtime": 389.9026,
-      "eval_samples_per_second": 5.63,
-      "eval_steps_per_second": 0.564,
+      "eval_loss": 5.626201629638672,
+      "eval_rouge1": 25.2705,
+      "eval_rouge2": 9.9865,
+      "eval_rougeL": 21.8678,
+      "eval_runtime": 372.5145,
+      "eval_samples_per_second": 5.892,
+      "eval_steps_per_second": 0.591,
       "step": 1528
     }
   ],
   "logging_steps": 500,
-  "max_steps": 15280,
+  "max_steps": 7640,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {

checkpoint-1528/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0647ffd63ecebccd487ca666ebff2e4007c4297f5565f476a9bab5a6aa5081d1
-size 6968
+oid sha256:0db0c5c28c9a1e67e13675dc0782cb65baca4516e1c39cf487f5c45dc85be680
+size 6904