Training in progress, step 40000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7378cec3d952e350bd0782f3f66d4685e17a260af33a26310893b255a4ef456f
+oid sha256:6a5f2c3a5f3243656fca6b8797830979cd9ddff00ba52ad64c6bfde0d1160a4c
 size 1668076741

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:faf2486a9aed6db196b4c0c2d69f6b9ce2fe47a70db1927026ee80ceceaa0b78
+oid sha256:e5ec64142d1f0a3cb201e869b089a93963530f2ebd5859646122ef979057409f
 size 834053717

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6577590482bd95dc8f5fc1b9185d2c76e77162ca88ae8b6a1adbd2a0c1f833f1
+oid sha256:deebbc10a778a0deea9d7ee66aff117a2ff1030819c46684fc2078337a13f7de
 size 17641

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcd693cfa252394680d67f0702bcf07e039830fa5cbdfd3fc37f2bc2c7ddcfe5
+oid sha256:1320d1dde11a646ceaddba0cc632bafb7ccddf41a813071e1e3284de4cbb041e
 size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb8a20482e53552d32fdf801d3bee556547de0b51c5f96c7d7318698a6a89bd8
+oid sha256:311fdf8544236bc9183022ce6f609475cd525084b757c5f4405c36c6da5663bf
 size 627

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.6,
-  "global_step": 30000,
+  "epoch": 0.8,
+  "global_step": 40000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -213,11 +213,80 @@
       "eval_samples_per_second": 27.394,
       "eval_steps_per_second": 0.685,
       "step": 30000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.0377012633509545e-05,
+      "loss": 0.8178,
+      "step": 31000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.444018944243678e-06,
+      "loss": 0.8298,
+      "step": 32000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 8.535320146484432e-06,
+      "loss": 0.8438,
+      "step": 33000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.655754766677537e-06,
+      "loss": 0.8532,
+      "step": 34000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 6.807408671366549e-06,
+      "loss": 0.8595,
+      "step": 35000
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 5.995680124886882e-06,
+      "loss": 0.8587,
+      "step": 36000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 5.222494120486821e-06,
+      "loss": 0.8702,
+      "step": 37000
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.492770657745851e-06,
+      "loss": 0.8598,
+      "step": 38000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.808240256977668e-06,
+      "loss": 0.8558,
+      "step": 39000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.1726485739618023e-06,
+      "loss": 0.8659,
+      "step": 40000
+    },
+    {
+      "epoch": 0.8,
+      "eval_accuracy": 0.8498886317841439,
+      "eval_loss": 0.6668724417686462,
+      "eval_runtime": 209.1326,
+      "eval_samples_per_second": 23.908,
+      "eval_steps_per_second": 0.598,
+      "step": 40000
     }
   ],
   "max_steps": 50000,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 9.469987651584e+18,
+  "total_flos": 1.2626650202112e+19,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:faf2486a9aed6db196b4c0c2d69f6b9ce2fe47a70db1927026ee80ceceaa0b78
+oid sha256:e5ec64142d1f0a3cb201e869b089a93963530f2ebd5859646122ef979057409f
 size 834053717