Training in progress, step 24000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09db5475ee35a12c8b710afb6e180ab5bfcd8ce48967fc58bd5457a4e3977985
+oid sha256:bfc5c32d531b4d03970087391e1f7ca6763b9d1de362d6cda421b7d124b979c8
 size 496262784

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff60aa4ce7613d8a03c55e84221468eaecbd9f2572db7e051cb5a7fa0b4c3a04
+oid sha256:baff61bb16847080902923a972f48196a5da070aa7a9295b3b6f30d9c6da65c5
 size 992621963

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4281625b7592c11b921e209e5b1131cddcd67ffe5764c2033c0ef0adb2d550bb
+oid sha256:707f3295d65fc2b08af5a95df451c0c1f230854850d6c8193dd0137a6c028a4e
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0231b5cc777a2809e149f15d91a081c8817fce35708f3c12c17c3012604a372
+oid sha256:acdbaf996d8220fb05cb901198d39146ebe82d7a748d641503b0d9a4f0d91bc3
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.44,
+  "epoch": 0.48,
   "eval_steps": 1000,
-  "global_step": 22000,
+  "global_step": 24000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1746,6 +1746,164 @@
       "eval_samples_per_second": 310.823,
       "eval_steps_per_second": 19.537,
       "step": 22000
+    },
+    {
+      "epoch": 0.442,
+      "grad_norm": 0.1982349008321762,
+      "learning_rate": 0.0003559357984104863,
+      "loss": 3.4337942504882815,
+      "step": 22100
+    },
+    {
+      "epoch": 0.444,
+      "grad_norm": 0.19478100538253784,
+      "learning_rate": 0.00035407722659740545,
+      "loss": 3.401671142578125,
+      "step": 22200
+    },
+    {
+      "epoch": 0.446,
+      "grad_norm": 0.20211219787597656,
+      "learning_rate": 0.0003522165070409252,
+      "loss": 3.416873779296875,
+      "step": 22300
+    },
+    {
+      "epoch": 0.448,
+      "grad_norm": 0.23854756355285645,
+      "learning_rate": 0.0003503537136418055,
+      "loss": 3.445356750488281,
+      "step": 22400
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.1983487904071808,
+      "learning_rate": 0.00034848892038317174,
+      "loss": 3.4137481689453124,
+      "step": 22500
+    },
+    {
+      "epoch": 0.452,
+      "grad_norm": 0.20169700682163239,
+      "learning_rate": 0.00034662220132757596,
+      "loss": 3.4005947875976563,
+      "step": 22600
+    },
+    {
+      "epoch": 0.454,
+      "grad_norm": 0.19934488832950592,
+      "learning_rate": 0.00034475363061405555,
+      "loss": 3.4056222534179685,
+      "step": 22700
+    },
+    {
+      "epoch": 0.456,
+      "grad_norm": 0.212734192609787,
+      "learning_rate": 0.00034288328245518906,
+      "loss": 3.399494323730469,
+      "step": 22800
+    },
+    {
+      "epoch": 0.458,
+      "grad_norm": 0.2278880625963211,
+      "learning_rate": 0.00034101123113414807,
+      "loss": 3.403822326660156,
+      "step": 22900
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.4838724434375763,
+      "learning_rate": 0.0003391375510017477,
+      "loss": 3.447413024902344,
+      "step": 23000
+    },
+    {
+      "epoch": 0.46,
+      "eval_accuracy": 0.3657575583429366,
+      "eval_loss": 3.438328742980957,
+      "eval_runtime": 6.1806,
+      "eval_samples_per_second": 314.047,
+      "eval_steps_per_second": 19.739,
+      "step": 23000
+    },
+    {
+      "epoch": 0.462,
+      "grad_norm": 0.4579889178276062,
+      "learning_rate": 0.00033726231647349304,
+      "loss": 3.4946719360351564,
+      "step": 23100
+    },
+    {
+      "epoch": 0.464,
+      "grad_norm": 0.724980890750885,
+      "learning_rate": 0.00033538560202662403,
+      "loss": 3.561537170410156,
+      "step": 23200
+    },
+    {
+      "epoch": 0.466,
+      "grad_norm": 1.940724492073059,
+      "learning_rate": 0.00033350748219715745,
+      "loss": 3.64361572265625,
+      "step": 23300
+    },
+    {
+      "epoch": 0.468,
+      "grad_norm": 2.4526569843292236,
+      "learning_rate": 0.00033162803157692656,
+      "loss": 3.734945983886719,
+      "step": 23400
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 2.5834765434265137,
+      "learning_rate": 0.0003297473248106184,
+      "loss": 3.86871826171875,
+      "step": 23500
+    },
+    {
+      "epoch": 0.472,
+      "grad_norm": 4.770744323730469,
+      "learning_rate": 0.0003278654365928096,
+      "loss": 3.974892883300781,
+      "step": 23600
+    },
+    {
+      "epoch": 0.474,
+      "grad_norm": 21.92555046081543,
+      "learning_rate": 0.0003259824416649996,
+      "loss": 4.260992431640625,
+      "step": 23700
+    },
+    {
+      "epoch": 0.476,
+      "grad_norm": 3082.5625,
+      "learning_rate": 0.00032409841481264223,
+      "loss": 5.928079833984375,
+      "step": 23800
+    },
+    {
+      "epoch": 0.478,
+      "grad_norm": 3319.344482421875,
+      "learning_rate": 0.0003222134308621751,
+      "loss": 9.030023803710938,
+      "step": 23900
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 322.2440490722656,
+      "learning_rate": 0.00032032756467804865,
+      "loss": 8.4674365234375,
+      "step": 24000
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.05309769310108071,
+      "eval_loss": 7.991781711578369,
+      "eval_runtime": 5.9674,
+      "eval_samples_per_second": 325.269,
+      "eval_steps_per_second": 20.445,
+      "step": 24000
     }
   ],
   "logging_steps": 100,
@@ -1765,7 +1923,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.8981096448e+17,
+  "total_flos": 7.5252105216e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null