Training in progress, step 1800, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:badfbe39f8402d520d6a259ad2deb9dd7e1f6dfb6b9e1a7169877f5ecc01c510
+oid sha256:206c62db0b89449568ebdd8c25cd017d55ecb7bf6df86bab1e451b3b3fd8f88e
 size 579748776

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4613374d2023ed2b9eeed79a32ef746ff946c02062fd7d42e828096eaa259890
+oid sha256:5a9af1204c1f4878c612726d1537329b4144d6b734182cd55d2c644c1a10e4b7
 size 1159627083

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71ed970edcf023c3a09dae82375e820aadbfb7a0861253f35a94fb8633387ad3
+oid sha256:5445b36fdb78ff76c2ef206d44397c5aad17f330414e24d03bab0de9bf0e9222
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef3ace7f288164d473a0c04358ab1c2df011edbd9de5a6bde6ef84f6a044d402
+oid sha256:108a89056f023268994da88aac029418224df4448adc2b0cb4d1083cb65ac2b0
 size 14917

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbe99df1134c19a237dcc24ad13c86b0959bc0c02f203c836365c9b3181e5558
+oid sha256:6d0f2c84615336a1d1664a4244642bd13601391ec93a4749b9c0e0cdc17433aa
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.75,
+  "epoch": 0.9,
   "eval_steps": 100,
-  "global_step": 1500,
+  "global_step": 1800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -563,6 +563,117 @@
       "eval_samples_per_second": 64.555,
       "eval_steps_per_second": 2.13,
       "step": 1500
+    },
+    {
+      "epoch": 0.7625,
+      "grad_norm": 2972.94580078125,
+      "learning_rate": 1.5464406188694174e-05,
+      "loss": 97.7231,
+      "step": 1525
+    },
+    {
+      "epoch": 0.775,
+      "grad_norm": 2009.724609375,
+      "learning_rate": 1.3960998405293963e-05,
+      "loss": 97.6005,
+      "step": 1550
+    },
+    {
+      "epoch": 0.7875,
+      "grad_norm": 955.8128051757812,
+      "learning_rate": 1.2522535338666486e-05,
+      "loss": 97.5096,
+      "step": 1575
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 1708.3966064453125,
+      "learning_rate": 1.1151609196484447e-05,
+      "loss": 97.562,
+      "step": 1600
+    },
+    {
+      "epoch": 0.8,
+      "eval_accuracy": 0.15179921134611404,
+      "eval_loss": 6.027195930480957,
+      "eval_runtime": 7.3648,
+      "eval_samples_per_second": 65.853,
+      "eval_steps_per_second": 2.172,
+      "step": 1600
+    },
+    {
+      "epoch": 0.8125,
+      "grad_norm": 249.3505096435547,
+      "learning_rate": 9.850690480317837e-06,
+      "loss": 97.4989,
+      "step": 1625
+    },
+    {
+      "epoch": 0.825,
+      "grad_norm": 493.7890930175781,
+      "learning_rate": 8.622123533623077e-06,
+      "loss": 97.3748,
+      "step": 1650
+    },
+    {
+      "epoch": 0.8375,
+      "grad_norm": 1148.812744140625,
+      "learning_rate": 7.468122317077786e-06,
+      "loss": 97.3686,
+      "step": 1675
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 719.5657958984375,
+      "learning_rate": 6.390766418873812e-06,
+      "loss": 97.3206,
+      "step": 1700
+    },
+    {
+      "epoch": 0.85,
+      "eval_accuracy": 0.15286253086201587,
+      "eval_loss": 6.017110824584961,
+      "eval_runtime": 7.5198,
+      "eval_samples_per_second": 64.496,
+      "eval_steps_per_second": 2.128,
+      "step": 1700
+    },
+    {
+      "epoch": 0.8625,
+      "grad_norm": 585.3778686523438,
+      "learning_rate": 5.391997307158919e-06,
+      "loss": 97.2298,
+      "step": 1725
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 2988.042236328125,
+      "learning_rate": 4.473614831379824e-06,
+      "loss": 97.108,
+      "step": 1750
+    },
+    {
+      "epoch": 0.8875,
+      "grad_norm": 3200.808837890625,
+      "learning_rate": 3.6372739788319843e-06,
+      "loss": 97.0287,
+      "step": 1775
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 263.6023254394531,
+      "learning_rate": 2.88448189226046e-06,
+      "loss": 96.8684,
+      "step": 1800
+    },
+    {
+      "epoch": 0.9,
+      "eval_accuracy": 0.15323817650296123,
+      "eval_loss": 6.016338348388672,
+      "eval_runtime": 14.8413,
+      "eval_samples_per_second": 32.679,
+      "eval_steps_per_second": 1.078,
+      "step": 1800
     }
   ],
   "logging_steps": 25,
@@ -582,7 +693,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.003336035729408e+18,
+  "total_flos": 1.2040032428752896e+18,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null