Training in progress, step 20000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91f77bd1ac2f3fafb2d2dff8abe3cf8b6a15fc5e2c266220cfd41101a08926f2
+oid sha256:0ba3c2b02c2181522825a50de924743eb11a5c8ec14f96218442ef3bf9c669ad
 size 579824888

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bd91aa34b703c596d190d43204798100682f3f8c6905d18b1c963356249d128
+oid sha256:683d6ef85df4768d6187df8f33dd7d019dd22439aeb22f78e278b0cdd0d17a92
 size 1159794763

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2118ba9622183c33aa1d1262712b3091d1f33a50e27c8636e086286e6c061b97
+oid sha256:5451e921407cd3b87f821062b7ea647b65f6d0ce5a09f45a2e5fe63f79c5afdc
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:803e8ad8cb04d8d95ab5fe814114135859d42a7ef5d29f0ea36f7faef8803122
+oid sha256:fdc6d60c7c4fe4f9e140eb1a01d69a125cb28ae84f348f2ce7bfdb028834e194
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.36,
+  "epoch": 0.4,
   "eval_steps": 1000,
-  "global_step": 18000,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1430,6 +1430,164 @@
       "eval_samples_per_second": 287.759,
       "eval_steps_per_second": 18.087,
       "step": 18000
+    },
+    {
+      "epoch": 0.362,
+      "grad_norm": 0.20023606717586517,
+      "learning_rate": 0.0007128041074339486,
+      "loss": 3.4777850341796874,
+      "step": 18100
+    },
+    {
+      "epoch": 0.364,
+      "grad_norm": 0.4378589689731598,
+      "learning_rate": 0.0007099484953400711,
+      "loss": 3.483775634765625,
+      "step": 18200
+    },
+    {
+      "epoch": 0.366,
+      "grad_norm": 0.15016524493694305,
+      "learning_rate": 0.0007070845448841241,
+      "loss": 3.471934509277344,
+      "step": 18300
+    },
+    {
+      "epoch": 0.368,
+      "grad_norm": 0.17178508639335632,
+      "learning_rate": 0.0007042123698114138,
+      "loss": 3.465678405761719,
+      "step": 18400
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.17568770051002502,
+      "learning_rate": 0.0007013320841938966,
+      "loss": 3.484932556152344,
+      "step": 18500
+    },
+    {
+      "epoch": 0.372,
+      "grad_norm": 0.12633204460144043,
+      "learning_rate": 0.0006984438024256499,
+      "loss": 3.4721084594726563,
+      "step": 18600
+    },
+    {
+      "epoch": 0.374,
+      "grad_norm": 0.13481932878494263,
+      "learning_rate": 0.0006955476392183275,
+      "loss": 3.441580505371094,
+      "step": 18700
+    },
+    {
+      "epoch": 0.376,
+      "grad_norm": 0.142787367105484,
+      "learning_rate": 0.0006926437095966044,
+      "loss": 3.472591552734375,
+      "step": 18800
+    },
+    {
+      "epoch": 0.378,
+      "grad_norm": 0.15913543105125427,
+      "learning_rate": 0.000689732128893608,
+      "loss": 3.4664480590820315,
+      "step": 18900
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.5166953802108765,
+      "learning_rate": 0.0006868130127463385,
+      "loss": 3.44828125,
+      "step": 19000
+    },
+    {
+      "epoch": 0.38,
+      "eval_accuracy": 0.36941738224793846,
+      "eval_loss": 3.4189138412475586,
+      "eval_runtime": 6.3949,
+      "eval_samples_per_second": 303.523,
+      "eval_steps_per_second": 19.078,
+      "step": 19000
+    },
+    {
+      "epoch": 0.382,
+      "grad_norm": 0.138035848736763,
+      "learning_rate": 0.0006838864770910745,
+      "loss": 3.4636468505859375,
+      "step": 19100
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 0.2230982631444931,
+      "learning_rate": 0.0006809526381587703,
+      "loss": 3.4643569946289063,
+      "step": 19200
+    },
+    {
+      "epoch": 0.386,
+      "grad_norm": 0.12826211750507355,
+      "learning_rate": 0.0006780116124704383,
+      "loss": 3.45958984375,
+      "step": 19300
+    },
+    {
+      "epoch": 0.388,
+      "grad_norm": 0.1421106457710266,
+      "learning_rate": 0.0006750635168325223,
+      "loss": 3.4553872680664064,
+      "step": 19400
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.14258891344070435,
+      "learning_rate": 0.000672108468332257,
+      "loss": 3.4486114501953127,
+      "step": 19500
+    },
+    {
+      "epoch": 0.392,
+      "grad_norm": 0.24561843276023865,
+      "learning_rate": 0.0006691465843330194,
+      "loss": 3.4674017333984377,
+      "step": 19600
+    },
+    {
+      "epoch": 0.394,
+      "grad_norm": 0.13396546244621277,
+      "learning_rate": 0.0006661779824696658,
+      "loss": 3.4596041870117187,
+      "step": 19700
+    },
+    {
+      "epoch": 0.396,
+      "grad_norm": 0.145228311419487,
+      "learning_rate": 0.000663202780643862,
+      "loss": 3.450461120605469,
+      "step": 19800
+    },
+    {
+      "epoch": 0.398,
+      "grad_norm": 0.1595899611711502,
+      "learning_rate": 0.0006602210970193982,
+      "loss": 3.4370431518554687,
+      "step": 19900
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.1425502896308899,
+      "learning_rate": 0.0006572330500174975,
+      "loss": 3.441788330078125,
+      "step": 20000
+    },
+    {
+      "epoch": 0.4,
+      "eval_accuracy": 0.3703691381064293,
+      "eval_loss": 3.4063475131988525,
+      "eval_runtime": 6.6969,
+      "eval_samples_per_second": 289.838,
+      "eval_steps_per_second": 18.218,
+      "step": 20000
     }
   ],
   "logging_steps": 100,
@@ -1449,7 +1607,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.0559295602688e+17,
+  "total_flos": 7.839921733632e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null