Model save

all_results.json

@@ -7,9 +7,9 @@
     "eval_steps_per_second": 0.334,
     "eval_wer": 1.0,
     "total_flos": 1.3476444758728704e+17,
-    "train_loss": 19.624227905273436,
-    "train_runtime": 37.1321,
+    "train_loss": 19.56191415786743,
+    "train_runtime": 92.9692,
     "train_samples": 128,
-    "train_samples_per_second": 17.236,
-    "train_steps_per_second": 0.269
+    "train_samples_per_second": 6.884,
+    "train_steps_per_second": 0.108
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:03187d4fdffd6bfd6668c712b8b34435c8b8fd1d873425a1d3179c6faaa1434c
+oid sha256:7ea93def4f37206b423024351165f6f76fb31abde5bd5c8aa0c1691f50e4ca3b
 size 1262067185

special_tokens_map.json

@@ -1 +1 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 5.0,
     "total_flos": 1.3476444758728704e+17,
-    "train_loss": 19.624227905273436,
-    "train_runtime": 37.1321,
+    "train_loss": 19.56191415786743,
+    "train_runtime": 92.9692,
     "train_samples": 128,
-    "train_samples_per_second": 17.236,
-    "train_steps_per_second": 0.269
+    "train_samples_per_second": 6.884,
+    "train_steps_per_second": 0.108
 }

trainer_state.json

@@ -7,14 +7,164 @@
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.5,
+      "learning_rate": 7.5e-05,
+      "loss": 22.283,
+      "step": 1
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 38.53455352783203,
+      "eval_runtime": 5.6686,
+      "eval_samples_per_second": 22.581,
+      "eval_steps_per_second": 0.353,
+      "eval_wer": 1.0,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.5e-05,
+      "loss": 22.622,
+      "step": 2
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 38.53455352783203,
+      "eval_runtime": 5.4852,
+      "eval_samples_per_second": 23.336,
+      "eval_steps_per_second": 0.365,
+      "eval_wer": 1.0,
+      "step": 2
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 6.75e-05,
+      "loss": 23.0656,
+      "step": 3
+    },
+    {
+      "epoch": 1.5,
+      "eval_loss": 37.73825454711914,
+      "eval_runtime": 5.4328,
+      "eval_samples_per_second": 23.561,
+      "eval_steps_per_second": 0.368,
+      "eval_wer": 1.0,
+      "step": 3
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.75e-05,
+      "loss": 20.8353,
+      "step": 4
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 37.73825454711914,
+      "eval_runtime": 5.7103,
+      "eval_samples_per_second": 22.416,
+      "eval_steps_per_second": 0.35,
+      "eval_wer": 1.0,
+      "step": 4
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 5.9999999999999995e-05,
+      "loss": 21.4977,
+      "step": 5
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 36.794700622558594,
+      "eval_runtime": 5.4096,
+      "eval_samples_per_second": 23.661,
+      "eval_steps_per_second": 0.37,
+      "eval_wer": 1.0,
+      "step": 5
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 5.2499999999999995e-05,
+      "loss": 20.1685,
+      "step": 6
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 35.34020233154297,
+      "eval_runtime": 5.6574,
+      "eval_samples_per_second": 22.625,
+      "eval_steps_per_second": 0.354,
+      "eval_wer": 1.0,
+      "step": 6
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 4.4999999999999996e-05,
+      "loss": 19.3566,
+      "step": 7
+    },
+    {
+      "epoch": 3.5,
+      "eval_loss": 33.572959899902344,
+      "eval_runtime": 5.6556,
+      "eval_samples_per_second": 22.632,
+      "eval_steps_per_second": 0.354,
+      "eval_wer": 1.0,
+      "step": 7
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.75e-05,
+      "loss": 15.8992,
+      "step": 8
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 30.588499069213867,
+      "eval_runtime": 5.3488,
+      "eval_samples_per_second": 23.93,
+      "eval_steps_per_second": 0.374,
+      "eval_wer": 1.0,
+      "step": 8
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 2.9999999999999997e-05,
+      "loss": 15.0607,
+      "step": 9
+    },
+    {
+      "epoch": 4.5,
+      "eval_loss": 27.82217788696289,
+      "eval_runtime": 5.6203,
+      "eval_samples_per_second": 22.774,
+      "eval_steps_per_second": 0.356,
+      "eval_wer": 1.0,
+      "step": 9
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.2499999999999998e-05,
+      "loss": 14.8305,
+      "step": 10
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 25.9698429107666,
+      "eval_runtime": 5.6638,
+      "eval_samples_per_second": 22.599,
+      "eval_steps_per_second": 0.353,
+      "eval_wer": 1.0,
+      "step": 10
+    },
     {
       "epoch": 5.0,
       "step": 10,
       "total_flos": 1.3476444758728704e+17,
-      "train_loss": 19.624227905273436,
-      "train_runtime": 37.1321,
-      "train_samples_per_second": 17.236,
-      "train_steps_per_second": 0.269
+      "train_loss": 19.56191415786743,
+      "train_runtime": 92.9692,
+      "train_samples_per_second": 6.884,
+      "train_steps_per_second": 0.108
     }
   ],
   "max_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75f86ca385bb82aeecbcfec2b42d4f24ce3dfaaa81f2eaa0da96c75d5dc881f6
+oid sha256:205b839c6738065c780db23902fe5de7ab8a5a6745c332ab60e9b1d434f7d4e6
 size 3055