Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:109a1ff562e60b4a3659de3f31e49413def647fff60dd801ec9ed38b39374255
+oid sha256:d0c8b974e314b98a2712368ee23c630ef3eb30ab969cdc8f8a7c3fba6de61d4f
 size 579824888

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8cdb8b10ce287047c72839d766120849b2d9aacf931ff44a1b8510116bdc7e6
+oid sha256:7158c9f9ccd1ba85502b2bb2c5f9ec3f583780fa992ace238229391c1e323e63
 size 1159794763

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0a81ce99965f5650998c3fdf0932f734f4f0b6b9b4f4bc0d128c8b5f65e9ca4
+oid sha256:6d920d35377b6218b0a182ca529dc7f210a3398080e4578011b7878234683794
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7ad50ba85de2f08235ac51873a9e6e650ceaf5d93913d1c853bc575c8052ec1
+oid sha256:7594d618d6ba19521c7abdc5d9c3d75c88e696a2ccaf3aad9e2ee70281ff7494
 size 1465

last-checkpoint/trainer_state.json

@@ -2,37 +2,177 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0004,
-  "eval_steps": 10,
-  "global_step": 20,
+  "epoch": 0.04,
+  "eval_steps": 1000,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0002,
-      "eval_accuracy": 0.043638611041376174,
-      "eval_loss": 9.762117385864258,
-      "eval_runtime": 6.9234,
-      "eval_samples_per_second": 280.355,
-      "eval_steps_per_second": 17.621,
-      "step": 10
-    },
-    {
-      "epoch": 0.0004,
-      "eval_accuracy": 0.06910715419957232,
-      "eval_loss": 8.825312614440918,
-      "eval_runtime": 6.4973,
-      "eval_samples_per_second": 298.738,
-      "eval_steps_per_second": 18.777,
-      "step": 20
+      "epoch": 0.002,
+      "grad_norm": 0.5271317362785339,
+      "learning_rate": 0.00066,
+      "loss": 7.779569091796875,
+      "step": 100
+    },
+    {
+      "epoch": 0.004,
+      "grad_norm": 0.6387752890586853,
+      "learning_rate": 0.0009999976160274871,
+      "loss": 6.24430908203125,
+      "step": 200
+    },
+    {
+      "epoch": 0.006,
+      "grad_norm": 0.42202863097190857,
+      "learning_rate": 0.000999977956590203,
+      "loss": 5.698394165039063,
+      "step": 300
+    },
+    {
+      "epoch": 0.008,
+      "grad_norm": 0.42532625794410706,
+      "learning_rate": 0.0009999384399149713,
+      "loss": 5.405557861328125,
+      "step": 400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4566122889518738,
+      "learning_rate": 0.0009998790675712448,
+      "loss": 5.189588623046875,
+      "step": 500
+    },
+    {
+      "epoch": 0.012,
+      "grad_norm": 0.4048032760620117,
+      "learning_rate": 0.000999799841917069,
+      "loss": 5.048131713867187,
+      "step": 600
+    },
+    {
+      "epoch": 0.014,
+      "grad_norm": 0.33951064944267273,
+      "learning_rate": 0.0009997007660989881,
+      "loss": 4.923414611816407,
+      "step": 700
+    },
+    {
+      "epoch": 0.016,
+      "grad_norm": 0.3192354440689087,
+      "learning_rate": 0.0009995818440519195,
+      "loss": 4.7589013671875,
+      "step": 800
+    },
+    {
+      "epoch": 0.018,
+      "grad_norm": 0.4026377201080322,
+      "learning_rate": 0.0009994430804989986,
+      "loss": 4.643058166503907,
+      "step": 900
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.3453679084777832,
+      "learning_rate": 0.0009992844809513896,
+      "loss": 4.579913330078125,
+      "step": 1000
+    },
+    {
+      "epoch": 0.02,
+      "eval_accuracy": 0.269652397386301,
+      "eval_loss": 4.505138874053955,
+      "eval_runtime": 6.375,
+      "eval_samples_per_second": 304.469,
+      "eval_steps_per_second": 19.137,
+      "step": 1000
+    },
+    {
+      "epoch": 0.022,
+      "grad_norm": 0.3148439824581146,
+      "learning_rate": 0.0009991060517080686,
+      "loss": 4.5075772094726565,
+      "step": 1100
+    },
+    {
+      "epoch": 0.024,
+      "grad_norm": 0.3168841600418091,
+      "learning_rate": 0.0009989077998555718,
+      "loss": 4.395697326660156,
+      "step": 1200
+    },
+    {
+      "epoch": 0.026,
+      "grad_norm": 0.277462899684906,
+      "learning_rate": 0.0009986897332677144,
+      "loss": 4.3901171875,
+      "step": 1300
+    },
+    {
+      "epoch": 0.028,
+      "grad_norm": 0.27992936968803406,
+      "learning_rate": 0.0009984518606052791,
+      "loss": 4.348124084472656,
+      "step": 1400
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.2656252980232239,
+      "learning_rate": 0.0009981941913156701,
+      "loss": 4.3058538818359375,
+      "step": 1500
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 0.36868757009506226,
+      "learning_rate": 0.0009979167356325394,
+      "loss": 4.28273681640625,
+      "step": 1600
+    },
+    {
+      "epoch": 0.034,
+      "grad_norm": 0.25333118438720703,
+      "learning_rate": 0.0009976195045753797,
+      "loss": 4.249351501464844,
+      "step": 1700
+    },
+    {
+      "epoch": 0.036,
+      "grad_norm": 0.2491893470287323,
+      "learning_rate": 0.0009973025099490876,
+      "loss": 4.22731689453125,
+      "step": 1800
+    },
+    {
+      "epoch": 0.038,
+      "grad_norm": 0.28424108028411865,
+      "learning_rate": 0.0009969657643434933,
+      "loss": 4.169223022460938,
+      "step": 1900
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.2666165232658386,
+      "learning_rate": 0.0009966092811328612,
+      "loss": 4.170299682617188,
+      "step": 2000
+    },
+    {
+      "epoch": 0.04,
+      "eval_accuracy": 0.3056205014664501,
+      "eval_loss": 4.103638648986816,
+      "eval_runtime": 6.566,
+      "eval_samples_per_second": 295.612,
+      "eval_steps_per_second": 18.58,
+      "step": 2000
     }
   ],
   "logging_steps": 100,
   "max_steps": 50000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
-  "save_steps": 20,
+  "save_steps": 2000,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -45,7 +185,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 783992173363200.0,
+  "total_flos": 7.839921733632e+16,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:873e1a4204ced4196d959a031f55bc9264b3965e59e7590a3c6d11d44d86d0dc
+oid sha256:017bba44c08e4ed15b103dc1408e362cb8183d9697dddf3065950e73396f8ef3
 size 5265