Training in progress, step 500, checkpoint

checkpoint-500/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6db92a12b77676b0343d62b007088b0a738bece00d0c41341971775bf642b185
+oid sha256:a87645bd1efb35bb7fd14f5822be84cd64461d1b04a2833049473d2b7dc00837
 size 19690328

checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7aeae66019a19277d7702df58a327e1efde9f30e994d526598b3f31685ac5c27
+oid sha256:055c8c212e6c1c013a906c384e74edb5b868b7d6e2f6e436bea2c50d15de6834
 size 38087162

checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd8652d2ed9fd8591d94e3ecdf652e2bbb2a82e873960bc84e210d60c0cfaf3f
+oid sha256:1538e52e854338dd2aae0af4e7fcc325258dc3e94ef565eb208fab88bd65300d
 size 14244

checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c9aec9a0bb79e6d56e4e5df00ca0d57e668ee525c9d1ed0e14c3fce618d34f1
+oid sha256:72b537d392fe9609b93120675b14e506ce729b31b676afdc372c208d28cf519d
 size 1064

checkpoint-500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.7192752361297607,
-  "best_model_checkpoint": "output/checkpoint-400",
-  "epoch": 2.5526483726866624,
+  "best_metric": 0.45803794264793396,
+  "best_model_checkpoint": "output/checkpoint-150",
+  "epoch": 0.40108292389451516,
   "eval_steps": 50,
   "global_step": 500,
   "is_hyper_param_search": false,
@@ -9,167 +9,167 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.01,
-      "learning_rate": 2.777777777777778e-07,
-      "loss": 9.5569,
+      "epoch": 0.0,
+      "learning_rate": 0.0,
+      "loss": 1.6453,
       "step": 1
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 1.25e-05,
-      "loss": 6.958,
+      "epoch": 0.04,
+      "learning_rate": 9.777777777777779e-06,
+      "loss": 1.3216,
       "step": 50
     },
     {
-      "epoch": 0.26,
-      "eval_accuracy": 0.42857142857142855,
-      "eval_loss": 6.581231117248535,
-      "eval_runtime": 28.2966,
-      "eval_samples_per_second": 6.184,
-      "eval_steps_per_second": 1.555,
+      "epoch": 0.04,
+      "eval_accuracy": 0.8057142857142857,
+      "eval_loss": 0.6264824271202087,
+      "eval_runtime": 28.8159,
+      "eval_samples_per_second": 6.073,
+      "eval_steps_per_second": 1.527,
       "step": 50
     },
     {
-      "epoch": 0.51,
-      "learning_rate": 2.6388888888888892e-05,
-      "loss": 1.6141,
+      "epoch": 0.08,
+      "learning_rate": 2.088888888888889e-05,
+      "loss": 0.918,
       "step": 100
     },
     {
-      "epoch": 0.51,
-      "eval_accuracy": 0.5028571428571429,
-      "eval_loss": 1.6640697717666626,
-      "eval_runtime": 28.1139,
-      "eval_samples_per_second": 6.225,
-      "eval_steps_per_second": 1.565,
+      "epoch": 0.08,
+      "eval_accuracy": 0.7885714285714286,
+      "eval_loss": 0.49068325757980347,
+      "eval_runtime": 26.1642,
+      "eval_samples_per_second": 6.689,
+      "eval_steps_per_second": 1.682,
       "step": 100
     },
     {
-      "epoch": 0.77,
-      "learning_rate": 4.027777777777778e-05,
-      "loss": 0.8533,
+      "epoch": 0.12,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.7219,
       "step": 150
     },
     {
-      "epoch": 0.77,
-      "eval_accuracy": 0.6342857142857142,
-      "eval_loss": 1.2631328105926514,
-      "eval_runtime": 26.205,
-      "eval_samples_per_second": 6.678,
-      "eval_steps_per_second": 1.679,
+      "epoch": 0.12,
+      "eval_accuracy": 0.7942857142857143,
+      "eval_loss": 0.45803794264793396,
+      "eval_runtime": 26.1308,
+      "eval_samples_per_second": 6.697,
+      "eval_steps_per_second": 1.684,
       "step": 150
     },
     {
-      "epoch": 1.02,
-      "learning_rate": 4.908536585365854e-05,
-      "loss": 0.4474,
+      "epoch": 0.16,
+      "learning_rate": 4.311111111111111e-05,
+      "loss": 0.6619,
       "step": 200
     },
     {
-      "epoch": 1.02,
-      "eval_accuracy": 0.7142857142857143,
-      "eval_loss": 0.9036799073219299,
-      "eval_runtime": 26.208,
-      "eval_samples_per_second": 6.677,
-      "eval_steps_per_second": 1.679,
+      "epoch": 0.16,
+      "eval_accuracy": 0.8285714285714286,
+      "eval_loss": 0.4676758646965027,
+      "eval_runtime": 26.1149,
+      "eval_samples_per_second": 6.701,
+      "eval_steps_per_second": 1.685,
       "step": 200
     },
     {
-      "epoch": 1.28,
-      "learning_rate": 4.603658536585366e-05,
-      "loss": 0.2996,
+      "epoch": 0.2,
+      "learning_rate": 4.906953966699315e-05,
+      "loss": 0.633,
       "step": 250
     },
     {
-      "epoch": 1.28,
-      "eval_accuracy": 0.6742857142857143,
-      "eval_loss": 1.1182247400283813,
-      "eval_runtime": 26.2071,
-      "eval_samples_per_second": 6.678,
-      "eval_steps_per_second": 1.679,
+      "epoch": 0.2,
+      "eval_accuracy": 0.8114285714285714,
+      "eval_loss": 0.4800652265548706,
+      "eval_runtime": 26.1161,
+      "eval_samples_per_second": 6.701,
+      "eval_steps_per_second": 1.685,
       "step": 250
     },
     {
-      "epoch": 1.53,
-      "learning_rate": 4.298780487804878e-05,
-      "loss": 0.2403,
+      "epoch": 0.24,
+      "learning_rate": 4.66209598432909e-05,
+      "loss": 0.6147,
       "step": 300
     },
     {
-      "epoch": 1.53,
-      "eval_accuracy": 0.72,
-      "eval_loss": 1.076162338256836,
-      "eval_runtime": 26.2142,
-      "eval_samples_per_second": 6.676,
-      "eval_steps_per_second": 1.678,
+      "epoch": 0.24,
+      "eval_accuracy": 0.7657142857142857,
+      "eval_loss": 0.5112562775611877,
+      "eval_runtime": 26.1127,
+      "eval_samples_per_second": 6.702,
+      "eval_steps_per_second": 1.685,
       "step": 300
     },
     {
-      "epoch": 1.79,
-      "learning_rate": 3.9939024390243905e-05,
-      "loss": 0.1982,
+      "epoch": 0.28,
+      "learning_rate": 4.4172380019588636e-05,
+      "loss": 0.6355,
       "step": 350
     },
     {
-      "epoch": 1.79,
-      "eval_accuracy": 0.6571428571428571,
-      "eval_loss": 1.4400347471237183,
-      "eval_runtime": 26.2187,
-      "eval_samples_per_second": 6.675,
-      "eval_steps_per_second": 1.678,
+      "epoch": 0.28,
+      "eval_accuracy": 0.7828571428571428,
+      "eval_loss": 0.5210281610488892,
+      "eval_runtime": 26.1048,
+      "eval_samples_per_second": 6.704,
+      "eval_steps_per_second": 1.686,
       "step": 350
     },
     {
-      "epoch": 2.04,
-      "learning_rate": 3.6890243902439025e-05,
-      "loss": 0.1926,
+      "epoch": 0.32,
+      "learning_rate": 4.1723800195886386e-05,
+      "loss": 0.593,
       "step": 400
     },
     {
-      "epoch": 2.04,
-      "eval_accuracy": 0.7942857142857143,
-      "eval_loss": 0.7192752361297607,
-      "eval_runtime": 26.2201,
-      "eval_samples_per_second": 6.674,
-      "eval_steps_per_second": 1.678,
+      "epoch": 0.32,
+      "eval_accuracy": 0.7885714285714286,
+      "eval_loss": 0.5098910927772522,
+      "eval_runtime": 26.1072,
+      "eval_samples_per_second": 6.703,
+      "eval_steps_per_second": 1.685,
       "step": 400
     },
     {
-      "epoch": 2.3,
-      "learning_rate": 3.384146341463415e-05,
-      "loss": 0.1258,
+      "epoch": 0.36,
+      "learning_rate": 3.927522037218414e-05,
+      "loss": 0.6047,
       "step": 450
     },
     {
-      "epoch": 2.3,
-      "eval_accuracy": 0.7657142857142857,
-      "eval_loss": 1.0054305791854858,
-      "eval_runtime": 26.2093,
-      "eval_samples_per_second": 6.677,
-      "eval_steps_per_second": 1.679,
+      "epoch": 0.36,
+      "eval_accuracy": 0.7314285714285714,
+      "eval_loss": 0.5602952837944031,
+      "eval_runtime": 26.1116,
+      "eval_samples_per_second": 6.702,
+      "eval_steps_per_second": 1.685,
       "step": 450
     },
     {
-      "epoch": 2.55,
-      "learning_rate": 3.079268292682927e-05,
-      "loss": 0.0857,
+      "epoch": 0.4,
+      "learning_rate": 3.682664054848188e-05,
+      "loss": 0.5661,
       "step": 500
     },
     {
-      "epoch": 2.55,
-      "eval_accuracy": 0.7828571428571428,
-      "eval_loss": 1.1228740215301514,
-      "eval_runtime": 26.1929,
-      "eval_samples_per_second": 6.681,
-      "eval_steps_per_second": 1.68,
+      "epoch": 0.4,
+      "eval_accuracy": 0.7542857142857143,
+      "eval_loss": 0.5670996904373169,
+      "eval_runtime": 26.1104,
+      "eval_samples_per_second": 6.702,
+      "eval_steps_per_second": 1.685,
       "step": 500
     }
   ],
   "logging_steps": 50,
-  "max_steps": 1000,
-  "num_train_epochs": 6,
+  "max_steps": 1246,
+  "num_train_epochs": 1,
   "save_steps": 50,
-  "total_flos": 1.863409147892736e+17,
+  "total_flos": 1.846344012865536e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a0ef36d7c31265f68a3edc8ae53ce9272dcc90a79f261e29a2a46e040c77a33
+oid sha256:2b7c1677f66b49d74d914a320393e2fa29530a9670d5824a30beac9590468d16
 size 4600