Training in progress, step 30000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c86c6ef1e9ed073386178347021ac781b8c02e77572c298de3cabb7139ff7bb
+oid sha256:7378cec3d952e350bd0782f3f66d4685e17a260af33a26310893b255a4ef456f
 size 1668076741

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:567827f0a9fb4e7e213e112d1698cfdc1ed84a2aa6f8e8ac3a3d6499893d619a
+oid sha256:faf2486a9aed6db196b4c0c2d69f6b9ce2fe47a70db1927026ee80ceceaa0b78
 size 834053717

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d83595397313a34341269e19cddb27d1911528d5cb71c1c4021a0f77255c067d
+oid sha256:6577590482bd95dc8f5fc1b9185d2c76e77162ca88ae8b6a1adbd2a0c1f833f1
 size 17641

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ec0071e4302006204e99cd369255e47065d131099e23d2fd8502b466b28f338
+oid sha256:fcd693cfa252394680d67f0702bcf07e039830fa5cbdfd3fc37f2bc2c7ddcfe5
 size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:843359b4ca576c0da051410b99ae90c5c724f142fad0be29b3fca9e912a14ce4
+oid sha256:cb8a20482e53552d32fdf801d3bee556547de0b51c5f96c7d7318698a6a89bd8
 size 627

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.4,
-  "global_step": 20000,
+  "epoch": 0.6,
+  "global_step": 30000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -144,11 +144,80 @@
       "eval_samples_per_second": 22.424,
       "eval_steps_per_second": 0.561,
       "step": 20000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 2.011372523790927e-05,
+      "loss": 0.8297,
+      "step": 21000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9171527138850477e-05,
+      "loss": 0.8453,
+      "step": 22000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.8210155373841292e-05,
+      "loss": 0.8384,
+      "step": 23000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7234746416166368e-05,
+      "loss": 0.828,
+      "step": 24000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.624956548101695e-05,
+      "loss": 0.8226,
+      "step": 25000
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.5259912447470205e-05,
+      "loss": 0.8188,
+      "step": 26000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.4268134252092541e-05,
+      "loss": 0.807,
+      "step": 27000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.3279556319416353e-05,
+      "loss": 0.8235,
+      "step": 28000
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.2298501449209877e-05,
+      "loss": 0.802,
+      "step": 29000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.1330221470267496e-05,
+      "loss": 0.8327,
+      "step": 30000
+    },
+    {
+      "epoch": 0.6,
+      "eval_accuracy": 0.8488932773912237,
+      "eval_loss": 0.6712204217910767,
+      "eval_runtime": 182.5232,
+      "eval_samples_per_second": 27.394,
+      "eval_steps_per_second": 0.685,
+      "step": 30000
     }
   ],
   "max_steps": 50000,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 6.313325101056e+18,
+  "total_flos": 9.469987651584e+18,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:567827f0a9fb4e7e213e112d1698cfdc1ed84a2aa6f8e8ac3a3d6499893d619a
+oid sha256:faf2486a9aed6db196b4c0c2d69f6b9ce2fe47a70db1927026ee80ceceaa0b78
 size 834053717