Training in progress, step 22000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80f6cba2576f1ee2628aa99fb4980b692c7f23dc4a1cc36b06da1ee070ed4750
+oid sha256:09db5475ee35a12c8b710afb6e180ab5bfcd8ce48967fc58bd5457a4e3977985
 size 496262784

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:238c57d596e4b481c83fa003f41489533d0e0d6fa65bc29cfff61620056f3f43
+oid sha256:ff60aa4ce7613d8a03c55e84221468eaecbd9f2572db7e051cb5a7fa0b4c3a04
 size 992621963

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2d3bdd032f891e90f3fae19dbb6eb0bfaf6ef591769711ba856b31e40586a5e
+oid sha256:4281625b7592c11b921e209e5b1131cddcd67ffe5764c2033c0ef0adb2d550bb
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c5b7f9f4f6c99c7a7ebbb7780b490f0a9830b3a79c1c5a07331e7e6edccccd28
+oid sha256:a0231b5cc777a2809e149f15d91a081c8817fce35708f3c12c17c3012604a372
 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": 329.872,
       "eval_steps_per_second": 20.734,
       "step": 20000
+    },
+    {
+      "epoch": 0.402,
+      "grad_norm": 0.20392677187919617,
+      "learning_rate": 0.0003925432549872678,
+      "loss": 3.467202453613281,
+      "step": 20100
+    },
+    {
+      "epoch": 0.404,
+      "grad_norm": 0.21144644916057587,
+      "learning_rate": 0.0003907430044951278,
+      "loss": 3.463043212890625,
+      "step": 20200
+    },
+    {
+      "epoch": 0.406,
+      "grad_norm": 0.1932135373353958,
+      "learning_rate": 0.00038893915003323544,
+      "loss": 3.484923095703125,
+      "step": 20300
+    },
+    {
+      "epoch": 0.408,
+      "grad_norm": 0.20958249270915985,
+      "learning_rate": 0.00038713176324388387,
+      "loss": 3.483151550292969,
+      "step": 20400
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.4303552210330963,
+      "learning_rate": 0.00038532091590965674,
+      "loss": 3.477305603027344,
+      "step": 20500
+    },
+    {
+      "epoch": 0.412,
+      "grad_norm": 0.20084762573242188,
+      "learning_rate": 0.0003835066799505776,
+      "loss": 3.47593994140625,
+      "step": 20600
+    },
+    {
+      "epoch": 0.414,
+      "grad_norm": 0.18663835525512695,
+      "learning_rate": 0.0003816891274212534,
+      "loss": 3.465316162109375,
+      "step": 20700
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 0.1979358196258545,
+      "learning_rate": 0.00037986833050801256,
+      "loss": 3.465248718261719,
+      "step": 20800
+    },
+    {
+      "epoch": 0.418,
+      "grad_norm": 0.20419807732105255,
+      "learning_rate": 0.0003780443615260386,
+      "loss": 3.4529428100585937,
+      "step": 20900
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.18685045838356018,
+      "learning_rate": 0.0003762172929164974,
+      "loss": 3.4456027221679686,
+      "step": 21000
+    },
+    {
+      "epoch": 0.42,
+      "eval_accuracy": 0.3708611474909034,
+      "eval_loss": 3.402836799621582,
+      "eval_runtime": 5.7802,
+      "eval_samples_per_second": 335.801,
+      "eval_steps_per_second": 21.106,
+      "step": 21000
+    },
+    {
+      "epoch": 0.422,
+      "grad_norm": 0.18441364169120789,
+      "learning_rate": 0.0003743871972436601,
+      "loss": 3.4683151245117188,
+      "step": 21100
+    },
+    {
+      "epoch": 0.424,
+      "grad_norm": 0.18499158322811127,
+      "learning_rate": 0.0003725541471920217,
+      "loss": 3.440291748046875,
+      "step": 21200
+    },
+    {
+      "epoch": 0.426,
+      "grad_norm": 0.19171582162380219,
+      "learning_rate": 0.00037071821556341393,
+      "loss": 3.4521221923828125,
+      "step": 21300
+    },
+    {
+      "epoch": 0.428,
+      "grad_norm": 0.2012597918510437,
+      "learning_rate": 0.0003688794752741139,
+      "loss": 3.4446597290039063,
+      "step": 21400
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.20872637629508972,
+      "learning_rate": 0.000367037999351948,
+      "loss": 3.4504824829101564,
+      "step": 21500
+    },
+    {
+      "epoch": 0.432,
+      "grad_norm": 0.19605720043182373,
+      "learning_rate": 0.0003651938609333918,
+      "loss": 3.432515869140625,
+      "step": 21600
+    },
+    {
+      "epoch": 0.434,
+      "grad_norm": 0.21077193319797516,
+      "learning_rate": 0.00036334713326066496,
+      "loss": 3.430672607421875,
+      "step": 21700
+    },
+    {
+      "epoch": 0.436,
+      "grad_norm": 0.20861658453941345,
+      "learning_rate": 0.00036149788967882265,
+      "loss": 3.434753112792969,
+      "step": 21800
+    },
+    {
+      "epoch": 0.438,
+      "grad_norm": 0.18320897221565247,
+      "learning_rate": 0.00035964620363284266,
+      "loss": 3.4054229736328123,
+      "step": 21900
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.20277726650238037,
+      "learning_rate": 0.00035779214866470794,
+      "loss": 3.412484130859375,
+      "step": 22000
+    },
+    {
+      "epoch": 0.44,
+      "eval_accuracy": 0.37140054302511166,
+      "eval_loss": 3.3950016498565674,
+      "eval_runtime": 6.2447,
+      "eval_samples_per_second": 310.823,
+      "eval_steps_per_second": 19.537,
+      "step": 22000
     }
   ],
   "logging_steps": 100,
@@ -1607,7 +1765,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.271008768e+17,
+  "total_flos": 6.8981096448e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null