Upload trainer_state.json

trainer_state.json

@@ -0,0 +1,233 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "global_step": 6250,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 9.0816,
+      "step": 400
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 4.077483177185059,
+      "eval_runtime": 9.5507,
+      "eval_samples_per_second": 13.297,
+      "eval_wer": 1.0,
+      "step": 400
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2e-05,
+      "loss": 3.6146,
+      "step": 800
+    },
+    {
+      "epoch": 1.28,
+      "eval_loss": 3.7597246170043945,
+      "eval_runtime": 9.6065,
+      "eval_samples_per_second": 13.22,
+      "eval_wer": 1.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2e-05,
+      "loss": 3.4563,
+      "step": 1200
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 3.7710764408111572,
+      "eval_runtime": 9.4925,
+      "eval_samples_per_second": 13.379,
+      "eval_wer": 1.0,
+      "step": 1200
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 2e-05,
+      "loss": 3.443,
+      "step": 1600
+    },
+    {
+      "epoch": 2.56,
+      "eval_loss": 3.6463630199432373,
+      "eval_runtime": 9.5724,
+      "eval_samples_per_second": 13.267,
+      "eval_wer": 1.0,
+      "step": 1600
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 2e-05,
+      "loss": 3.0853,
+      "step": 2000
+    },
+    {
+      "epoch": 3.2,
+      "eval_loss": 2.549088716506958,
+      "eval_runtime": 9.6262,
+      "eval_samples_per_second": 13.193,
+      "eval_wer": 1.005800464037123,
+      "step": 2000
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 2e-05,
+      "loss": 1.9334,
+      "step": 2400
+    },
+    {
+      "epoch": 3.84,
+      "eval_loss": 1.648258924484253,
+      "eval_runtime": 9.5891,
+      "eval_samples_per_second": 13.244,
+      "eval_wer": 0.7807424593967517,
+      "step": 2400
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 2e-05,
+      "loss": 1.3425,
+      "step": 2800
+    },
+    {
+      "epoch": 4.48,
+      "eval_loss": 1.4189220666885376,
+      "eval_runtime": 9.5676,
+      "eval_samples_per_second": 13.274,
+      "eval_wer": 0.722737819025522,
+      "step": 2800
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 2e-05,
+      "loss": 1.0799,
+      "step": 3200
+    },
+    {
+      "epoch": 5.12,
+      "eval_loss": 1.2596975564956665,
+      "eval_runtime": 9.5729,
+      "eval_samples_per_second": 13.267,
+      "eval_wer": 0.6310904872389791,
+      "step": 3200
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 2e-05,
+      "loss": 0.9505,
+      "step": 3600
+    },
+    {
+      "epoch": 5.76,
+      "eval_loss": 1.205084204673767,
+      "eval_runtime": 9.5232,
+      "eval_samples_per_second": 13.336,
+      "eval_wer": 0.5974477958236659,
+      "step": 3600
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 2e-05,
+      "loss": 0.8179,
+      "step": 4000
+    },
+    {
+      "epoch": 6.4,
+      "eval_loss": 1.1981126070022583,
+      "eval_runtime": 9.5515,
+      "eval_samples_per_second": 13.296,
+      "eval_wer": 0.5974477958236659,
+      "step": 4000
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 2e-05,
+      "loss": 0.7435,
+      "step": 4400
+    },
+    {
+      "epoch": 7.04,
+      "eval_loss": 1.194669246673584,
+      "eval_runtime": 9.5839,
+      "eval_samples_per_second": 13.251,
+      "eval_wer": 0.5696055684454756,
+      "step": 4400
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 2e-05,
+      "loss": 0.7038,
+      "step": 4800
+    },
+    {
+      "epoch": 7.68,
+      "eval_loss": 1.185306191444397,
+      "eval_runtime": 9.5858,
+      "eval_samples_per_second": 13.249,
+      "eval_wer": 0.5603248259860789,
+      "step": 4800
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2e-05,
+      "loss": 0.6528,
+      "step": 5200
+    },
+    {
+      "epoch": 8.32,
+      "eval_loss": 1.193933367729187,
+      "eval_runtime": 9.6168,
+      "eval_samples_per_second": 13.206,
+      "eval_wer": 0.5556844547563805,
+      "step": 5200
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 2e-05,
+      "loss": 0.5934,
+      "step": 5600
+    },
+    {
+      "epoch": 8.96,
+      "eval_loss": 1.1850134134292603,
+      "eval_runtime": 9.5808,
+      "eval_samples_per_second": 13.256,
+      "eval_wer": 0.5533642691415314,
+      "step": 5600
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 2e-05,
+      "loss": 0.5604,
+      "step": 6000
+    },
+    {
+      "epoch": 9.6,
+      "eval_loss": 1.1631267070770264,
+      "eval_runtime": 9.6116,
+      "eval_samples_per_second": 13.213,
+      "eval_wer": 0.537122969837587,
+      "step": 6000
+    },
+    {
+      "epoch": 10.0,
+      "step": 6250,
+      "total_flos": 3.666955074249646e+18,
+      "train_runtime": 7297.1898,
+      "train_samples_per_second": 0.856
+    }
+  ],
+  "max_steps": 6250,
+  "num_train_epochs": 10,
+  "total_flos": 3.666955074249646e+18,
+  "trial_name": null,
+  "trial_params": null
+}