Training in progress, step 14000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c622298879cc833d1bdcc0a29bd5f4d753b0f9b392bcfe41b3a3720e37227629
+oid sha256:2e0ff05e0e5d852455b4d83922c7abbecdec428e19e46c0b823ee661ac50cdb5
 size 496262784

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f770627df18e9419658687702329dfc460bbc5339dfe15ca4d75788aa1849e5a
+oid sha256:32d116e4c03c1c5f0b5164cbe30850c283753e36a9119b42ffefe95697cc08e3
 size 992621963

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e52021f1e1b36f10ce9c865a52c47d6a63c888218cae64f20627598071e5030f
+oid sha256:74ba315e55e014287fc4f548e786e1c64991f07f2dd31f970e2f14eab773946e
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:382f69ae5e5420217ac9e6a070a414359d614c97fe4b4d34a11abdd073f1e7c2
+oid sha256:238cbdda4406a9b9e4f702cada921a66b6ebd035a4f580dabf57521e3dc0c97d
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.24,
+  "epoch": 0.28,
   "eval_steps": 1000,
-  "global_step": 12000,
+  "global_step": 14000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -956,6 +956,164 @@
       "eval_samples_per_second": 342.004,
       "eval_steps_per_second": 21.496,
       "step": 12000
+    },
+    {
+      "epoch": 0.242,
+      "grad_norm": 0.19794468581676483,
+      "learning_rate": 0.0005188846240195041,
+      "loss": 3.563157958984375,
+      "step": 12100
+    },
+    {
+      "epoch": 0.244,
+      "grad_norm": 0.22670620679855347,
+      "learning_rate": 0.0005175873657931715,
+      "loss": 3.573887939453125,
+      "step": 12200
+    },
+    {
+      "epoch": 0.246,
+      "grad_norm": 0.2241670936346054,
+      "learning_rate": 0.0005162814658176576,
+      "loss": 3.614027099609375,
+      "step": 12300
+    },
+    {
+      "epoch": 0.248,
+      "grad_norm": 0.23707276582717896,
+      "learning_rate": 0.000514966975958382,
+      "loss": 3.628121643066406,
+      "step": 12400
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.22935494780540466,
+      "learning_rate": 0.0005136439484219223,
+      "loss": 3.6195343017578123,
+      "step": 12500
+    },
+    {
+      "epoch": 0.252,
+      "grad_norm": 0.21692432463169098,
+      "learning_rate": 0.0005123124357539403,
+      "loss": 3.62494140625,
+      "step": 12600
+    },
+    {
+      "epoch": 0.254,
+      "grad_norm": 0.1959449201822281,
+      "learning_rate": 0.0005109724908370951,
+      "loss": 3.619110412597656,
+      "step": 12700
+    },
+    {
+      "epoch": 0.256,
+      "grad_norm": 0.22109989821910858,
+      "learning_rate": 0.000509624166888943,
+      "loss": 3.609862060546875,
+      "step": 12800
+    },
+    {
+      "epoch": 0.258,
+      "grad_norm": 0.20637010037899017,
+      "learning_rate": 0.0005082675174598238,
+      "loss": 3.6038763427734377,
+      "step": 12900
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.21297694742679596,
+      "learning_rate": 0.000506902596430734,
+      "loss": 3.627516174316406,
+      "step": 13000
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.35723510890244603,
+      "eval_loss": 3.5337636470794678,
+      "eval_runtime": 5.6987,
+      "eval_samples_per_second": 340.603,
+      "eval_steps_per_second": 21.408,
+      "step": 13000
+    },
+    {
+      "epoch": 0.262,
+      "grad_norm": 0.2433704286813736,
+      "learning_rate": 0.0005055294580111867,
+      "loss": 3.600766296386719,
+      "step": 13100
+    },
+    {
+      "epoch": 0.264,
+      "grad_norm": 0.20984166860580444,
+      "learning_rate": 0.0005041481567370593,
+      "loss": 3.57820556640625,
+      "step": 13200
+    },
+    {
+      "epoch": 0.266,
+      "grad_norm": 0.22343853116035461,
+      "learning_rate": 0.0005027587474684261,
+      "loss": 3.603807373046875,
+      "step": 13300
+    },
+    {
+      "epoch": 0.268,
+      "grad_norm": 0.2158922404050827,
+      "learning_rate": 0.0005013612853873809,
+      "loss": 3.5946688842773438,
+      "step": 13400
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.21097783744335175,
+      "learning_rate": 0.0004999558259958449,
+      "loss": 3.589884033203125,
+      "step": 13500
+    },
+    {
+      "epoch": 0.272,
+      "grad_norm": 0.28342482447624207,
+      "learning_rate": 0.0004985424251133622,
+      "loss": 3.5845135498046874,
+      "step": 13600
+    },
+    {
+      "epoch": 0.274,
+      "grad_norm": 0.22716180980205536,
+      "learning_rate": 0.0004971211388748826,
+      "loss": 3.601893310546875,
+      "step": 13700
+    },
+    {
+      "epoch": 0.276,
+      "grad_norm": 0.2315467894077301,
+      "learning_rate": 0.000495692023728533,
+      "loss": 3.583456726074219,
+      "step": 13800
+    },
+    {
+      "epoch": 0.278,
+      "grad_norm": 0.20143155753612518,
+      "learning_rate": 0.0004942551364333746,
+      "loss": 3.5840383911132814,
+      "step": 13900
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.2158096581697464,
+      "learning_rate": 0.0004928105340571493,
+      "loss": 3.5770263671875,
+      "step": 14000
+    },
+    {
+      "epoch": 0.28,
+      "eval_accuracy": 0.3599905630986912,
+      "eval_loss": 3.508211851119995,
+      "eval_runtime": 9.9993,
+      "eval_samples_per_second": 194.114,
+      "eval_steps_per_second": 12.201,
+      "step": 14000
     }
   ],
   "logging_steps": 100,
@@ -975,7 +1133,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.7626052608e+17,
+  "total_flos": 4.3897061376e+17,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null