Training in progress, step 22000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ba3c2b02c2181522825a50de924743eb11a5c8ec14f96218442ef3bf9c669ad
+oid sha256:e851d31aa87e3f44cdb6b12860dfca56948ed93c4c3e52e3a3bc7173fa578a8b
 size 579824888

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:683d6ef85df4768d6187df8f33dd7d019dd22439aeb22f78e278b0cdd0d17a92
+oid sha256:df630ecbd687f1359a191cc034b3009e41a878d4ab7fdb04cf787f0bc23ecb0c
 size 1159794763

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5451e921407cd3b87f821062b7ea647b65f6d0ce5a09f45a2e5fe63f79c5afdc
+oid sha256:43b610cc94e504b24cb2bb3bb992068bab81d217de2c8ad61ad01dfb75e98503
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdc6d60c7c4fe4f9e140eb1a01d69a125cb28ae84f348f2ce7bfdb028834e194
+oid sha256:f23bcbe880608cbeb7ee4f55362f4ac1ed7848d0f02874b239279ca55eb68fab
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.4,
+  "epoch": 0.44,
   "eval_steps": 1000,
-  "global_step": 20000,
+  "global_step": 22000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1588,6 +1588,164 @@
       "eval_samples_per_second": 289.838,
       "eval_steps_per_second": 18.218,
       "step": 20000
+    },
+    {
+      "epoch": 0.402,
+      "grad_norm": 0.1547059416770935,
+      "learning_rate": 0.0006542387583121131,
+      "loss": 3.441893310546875,
+      "step": 20100
+    },
+    {
+      "epoch": 0.404,
+      "grad_norm": 0.15452757477760315,
+      "learning_rate": 0.000651238340825213,
+      "loss": 3.434819641113281,
+      "step": 20200
+    },
+    {
+      "epoch": 0.406,
+      "grad_norm": 0.13765129446983337,
+      "learning_rate": 0.0006482319167220591,
+      "loss": 3.4564175415039062,
+      "step": 20300
+    },
+    {
+      "epoch": 0.408,
+      "grad_norm": 0.15758726000785828,
+      "learning_rate": 0.0006452196054064732,
+      "loss": 3.4555868530273437,
+      "step": 20400
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.3239542543888092,
+      "learning_rate": 0.0006422015265160946,
+      "loss": 3.4481591796875,
+      "step": 20500
+    },
+    {
+      "epoch": 0.412,
+      "grad_norm": 0.15911568701267242,
+      "learning_rate": 0.0006391777999176294,
+      "loss": 3.44850341796875,
+      "step": 20600
+    },
+    {
+      "epoch": 0.414,
+      "grad_norm": 0.12856774032115936,
+      "learning_rate": 0.000636148545702089,
+      "loss": 3.438498840332031,
+      "step": 20700
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 0.17129512131214142,
+      "learning_rate": 0.000633113884180021,
+      "loss": 3.438189697265625,
+      "step": 20800
+    },
+    {
+      "epoch": 0.418,
+      "grad_norm": 0.14792455732822418,
+      "learning_rate": 0.0006300739358767311,
+      "loss": 3.425160217285156,
+      "step": 20900
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.13070790469646454,
+      "learning_rate": 0.0006270288215274957,
+      "loss": 3.4182763671875,
+      "step": 21000
+    },
+    {
+      "epoch": 0.42,
+      "eval_accuracy": 0.3719621193102593,
+      "eval_loss": 3.395620584487915,
+      "eval_runtime": 6.7127,
+      "eval_samples_per_second": 289.155,
+      "eval_steps_per_second": 18.175,
+      "step": 21000
+    },
+    {
+      "epoch": 0.422,
+      "grad_norm": 0.1333521008491516,
+      "learning_rate": 0.0006239786620727669,
+      "loss": 3.440789794921875,
+      "step": 21100
+    },
+    {
+      "epoch": 0.424,
+      "grad_norm": 0.14673130214214325,
+      "learning_rate": 0.0006209235786533696,
+      "loss": 3.41100341796875,
+      "step": 21200
+    },
+    {
+      "epoch": 0.426,
+      "grad_norm": 0.13843247294425964,
+      "learning_rate": 0.00061786369260569,
+      "loss": 3.4241146850585937,
+      "step": 21300
+    },
+    {
+      "epoch": 0.428,
+      "grad_norm": 0.15212669968605042,
+      "learning_rate": 0.0006147991254568566,
+      "loss": 3.4158154296875,
+      "step": 21400
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.1734033226966858,
+      "learning_rate": 0.0006117299989199135,
+      "loss": 3.4223089599609375,
+      "step": 21500
+    },
+    {
+      "epoch": 0.432,
+      "grad_norm": 0.15912015736103058,
+      "learning_rate": 0.0006086564348889863,
+      "loss": 3.405228271484375,
+      "step": 21600
+    },
+    {
+      "epoch": 0.434,
+      "grad_norm": 0.1569402515888214,
+      "learning_rate": 0.0006055785554344417,
+      "loss": 3.4033087158203124,
+      "step": 21700
+    },
+    {
+      "epoch": 0.436,
+      "grad_norm": 0.15008965134620667,
+      "learning_rate": 0.0006024964827980378,
+      "loss": 3.4061663818359373,
+      "step": 21800
+    },
+    {
+      "epoch": 0.438,
+      "grad_norm": 0.12902535498142242,
+      "learning_rate": 0.0005994103393880712,
+      "loss": 3.3769439697265624,
+      "step": 21900
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.14630058407783508,
+      "learning_rate": 0.0005963202477745134,
+      "loss": 3.3858111572265623,
+      "step": 22000
+    },
+    {
+      "epoch": 0.44,
+      "eval_accuracy": 0.372494457332805,
+      "eval_loss": 3.3869516849517822,
+      "eval_runtime": 6.5171,
+      "eval_samples_per_second": 297.832,
+      "eval_steps_per_second": 18.72,
+      "step": 22000
     }
   ],
   "logging_steps": 100,
@@ -1607,7 +1765,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.839921733632e+17,
+  "total_flos": 8.6239139069952e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null