Training in progress, step 400

checkpoint-100/trainer_state.json

@@ -1,55 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 1.098360655737705,
-  "global_step": 100,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.22,
-      "learning_rate": 2.5e-06,
-      "loss": 3.5867,
-      "step": 20
-    },
-    {
-      "epoch": 0.44,
-      "learning_rate": 5e-06,
-      "loss": 3.5457,
-      "step": 40
-    },
-    {
-      "epoch": 0.66,
-      "learning_rate": 7.5e-06,
-      "loss": 3.4513,
-      "step": 60
-    },
-    {
-      "epoch": 0.87,
-      "learning_rate": 1e-05,
-      "loss": 3.3432,
-      "step": 80
-    },
-    {
-      "epoch": 1.1,
-      "learning_rate": 1.25e-05,
-      "loss": 3.3533,
-      "step": 100
-    },
-    {
-      "epoch": 1.1,
-      "eval_loss": 3.2806732654571533,
-      "eval_runtime": 190.4728,
-      "eval_samples_per_second": 25.426,
-      "eval_steps_per_second": 0.798,
-      "eval_wer": 1.0,
-      "step": 100
-    }
-  ],
-  "max_steps": 4550,
-  "num_train_epochs": 50,
-  "total_flos": 1.5920272455652972e+18,
-  "trial_name": null,
-  "trial_params": null
-}

{checkpoint-100 → checkpoint-400}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfe9830bfa01dc750a3e9885c7197d90f6d28a575aa94090eb3ff702d43b1eb4
-size 2490337361
+oid sha256:b573eee933c6c36bd4ddb6906221d3843d7e3e13dab0b96a39404d3a30f74898
+size 2490337809

{checkpoint-100 → checkpoint-400}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36716972ea06ded09ef95dae3d4c1c67df97e8b98994fe6df6c3873c9e26f6d2
+oid sha256:28c7e3ec36d37a20298023cfa6ac730ae02e82d170ef8859ec5740936e2bc809
 size 1262063089

{checkpoint-100 → checkpoint-400}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65a9e117b12baf17673014db5778f9eb47486ba7e811d1bf9e31465235ab3385
-size 14567
+oid sha256:855318e4e46176f35f66cfa7dbcef4ac9632248b73c3328855184518a3b76d7f
+size 14503

{checkpoint-100 → checkpoint-400}/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13a3423b2fe42f204bc8fe2c666ff379f9fd753a0f13613064a5e71e86b519e8
+oid sha256:476e510c8ea7edbd2b51d1e76a4e037820a5639381c0d8b5d32dafa492795a1e
 size 559

{checkpoint-100 → checkpoint-400}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff74e92858f017aa05365d0a2e3c9e1d9c95680d66d2c770074de1484046e8f2
+oid sha256:8d06447f3e192636cdb71ae042312281cde114831b8524799c0556275e17d331
 size 623

checkpoint-400/trainer_state.json

@@ -0,0 +1,172 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.39344262295082,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.22,
+      "learning_rate": 2.5e-06,
+      "loss": 3.5867,
+      "step": 20
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 5e-06,
+      "loss": 3.5457,
+      "step": 40
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 7.5e-06,
+      "loss": 3.4513,
+      "step": 60
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 3.3432,
+      "step": 80
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.25e-05,
+      "loss": 3.3533,
+      "step": 100
+    },
+    {
+      "epoch": 1.1,
+      "eval_loss": 3.2806732654571533,
+      "eval_runtime": 190.4728,
+      "eval_samples_per_second": 25.426,
+      "eval_steps_per_second": 0.798,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.5e-05,
+      "loss": 3.2217,
+      "step": 120
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 3.1765,
+      "step": 140
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 2e-05,
+      "loss": 3.1408,
+      "step": 160
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.2499999999999998e-05,
+      "loss": 3.1165,
+      "step": 180
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.5e-05,
+      "loss": 3.1709,
+      "step": 200
+    },
+    {
+      "epoch": 2.2,
+      "eval_loss": 3.1325438022613525,
+      "eval_runtime": 192.4978,
+      "eval_samples_per_second": 25.159,
+      "eval_steps_per_second": 0.79,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 2.75e-05,
+      "loss": 3.079,
+      "step": 220
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 3e-05,
+      "loss": 3.0677,
+      "step": 240
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 3.2500000000000004e-05,
+      "loss": 3.0656,
+      "step": 260
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 3.5000000000000004e-05,
+      "loss": 3.1463,
+      "step": 280
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 3.75e-05,
+      "loss": 3.0573,
+      "step": 300
+    },
+    {
+      "epoch": 3.3,
+      "eval_loss": 3.0614514350891113,
+      "eval_runtime": 194.36,
+      "eval_samples_per_second": 24.918,
+      "eval_steps_per_second": 0.782,
+      "eval_wer": 1.0,
+      "step": 300
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 4e-05,
+      "loss": 3.0511,
+      "step": 320
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 4.25e-05,
+      "loss": 3.0358,
+      "step": 340
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 4.4999999999999996e-05,
+      "loss": 3.0416,
+      "step": 360
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 4.75e-05,
+      "loss": 3.1053,
+      "step": 380
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 5e-05,
+      "loss": 3.0314,
+      "step": 400
+    },
+    {
+      "epoch": 4.39,
+      "eval_loss": 3.0990231037139893,
+      "eval_runtime": 198.3688,
+      "eval_samples_per_second": 24.414,
+      "eval_steps_per_second": 0.766,
+      "eval_wer": 1.0,
+      "step": 400
+    }
+  ],
+  "max_steps": 4550,
+  "num_train_epochs": 50,
+  "total_flos": 6.222452805866187e+18,
+  "trial_name": null,
+  "trial_params": null
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:434bb79de859e935ca1aaf8411c1439a4e24ef8e2396b088b07525e56e0847d8
+oid sha256:28c7e3ec36d37a20298023cfa6ac730ae02e82d170ef8859ec5740936e2bc809
 size 1262063089