Training in progress, step 4000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:290d3415f88d0aa6d2afc82b9cd5f52a66da76a1b8dcef2153a477c366098b63
+oid sha256:8b8c4f09e4df38c2b9a7b4ea1bb4b3bb55ce4048cf496c490994769c5b20d4a8
 size 1442699

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32fe97d8864d635e74e59a243935c32329fbca3a7eff55bcfa48502a005f0d95
+oid sha256:8be8ea274e8c78fb1a6aed38fc7b39f6a706b7c0fc47d551c8c41142e7ae8382
 size 499774435

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87cfd50ff3a42065a545f46e5e0d163da023498a41ad02da368f633d32b47d2c
+oid sha256:af971d84c55a6494396ed13d3bc14bba9e161e3dc5123551d9be42796da1c2be
 size 14455

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b716fb1dcbc77ddc6a0ae7c4c98d559e36f1986fff006ef6982bd75247833235
+oid sha256:76cf98a8cde877f1e49a9e4b884c4dce3d51bb4035429d464e74acc35ddca024
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.04,
+  "epoch": 0.08,
   "eval_steps": 1000,
-  "global_step": 2000,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -166,6 +166,164 @@
       "eval_samples_per_second": 147.345,
       "eval_steps_per_second": 18.418,
       "step": 2000
+    },
+    {
+      "epoch": 0.042,
+      "grad_norm": 0.6110609173774719,
+      "learning_rate": 0.0016792,
+      "loss": 5.67332275390625,
+      "step": 2100
+    },
+    {
+      "epoch": 0.044,
+      "grad_norm": 0.579617977142334,
+      "learning_rate": 0.0017592,
+      "loss": 5.706051025390625,
+      "step": 2200
+    },
+    {
+      "epoch": 0.046,
+      "grad_norm": 0.5633976459503174,
+      "learning_rate": 0.0018392,
+      "loss": 5.6666259765625,
+      "step": 2300
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 0.6478831768035889,
+      "learning_rate": 0.0019192,
+      "loss": 5.682818603515625,
+      "step": 2400
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.8043723702430725,
+      "learning_rate": 0.0019992,
+      "loss": 5.635286254882812,
+      "step": 2500
+    },
+    {
+      "epoch": 0.052,
+      "grad_norm": 0.554682195186615,
+      "learning_rate": 0.001999978563623903,
+      "loss": 5.5472607421875,
+      "step": 2600
+    },
+    {
+      "epoch": 0.054,
+      "grad_norm": 0.6824977397918701,
+      "learning_rate": 0.0019999133871331223,
+      "loss": 5.5920654296875,
+      "step": 2700
+    },
+    {
+      "epoch": 0.056,
+      "grad_norm": 0.40844905376434326,
+      "learning_rate": 0.0019998044711915177,
+      "loss": 5.528501586914063,
+      "step": 2800
+    },
+    {
+      "epoch": 0.058,
+      "grad_norm": 0.6135262250900269,
+      "learning_rate": 0.0019996518205634257,
+      "loss": 5.487052612304687,
+      "step": 2900
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.4022127687931061,
+      "learning_rate": 0.0019994554419262797,
+      "loss": 5.5177392578125,
+      "step": 3000
+    },
+    {
+      "epoch": 0.06,
+      "eval_accuracy": 0.18825440313111547,
+      "eval_loss": 5.509125232696533,
+      "eval_runtime": 6.779,
+      "eval_samples_per_second": 147.515,
+      "eval_steps_per_second": 18.439,
+      "step": 3000
+    },
+    {
+      "epoch": 0.062,
+      "grad_norm": 0.7171342968940735,
+      "learning_rate": 0.001999215343870317,
+      "loss": 5.417413330078125,
+      "step": 3100
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 0.4542510509490967,
+      "learning_rate": 0.0019989315368982054,
+      "loss": 5.4095831298828125,
+      "step": 3200
+    },
+    {
+      "epoch": 0.066,
+      "grad_norm": 0.5080035924911499,
+      "learning_rate": 0.0019986040334245797,
+      "loss": 5.417955932617187,
+      "step": 3300
+    },
+    {
+      "epoch": 0.068,
+      "grad_norm": 0.3587779700756073,
+      "learning_rate": 0.001998232847775504,
+      "loss": 5.3721466064453125,
+      "step": 3400
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.45940911769866943,
+      "learning_rate": 0.0019978179961878404,
+      "loss": 5.320546875,
+      "step": 3500
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 0.3420865535736084,
+      "learning_rate": 0.001997359496808541,
+      "loss": 5.28208984375,
+      "step": 3600
+    },
+    {
+      "epoch": 0.074,
+      "grad_norm": 0.42247381806373596,
+      "learning_rate": 0.001996857369693855,
+      "loss": 5.29147216796875,
+      "step": 3700
+    },
+    {
+      "epoch": 0.076,
+      "grad_norm": 0.3335079550743103,
+      "learning_rate": 0.0019963116368084486,
+      "loss": 5.352775268554687,
+      "step": 3800
+    },
+    {
+      "epoch": 0.078,
+      "grad_norm": 0.6875544786453247,
+      "learning_rate": 0.001995722322024446,
+      "loss": 5.303611450195312,
+      "step": 3900
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.3652666211128235,
+      "learning_rate": 0.001995089451120385,
+      "loss": 5.26600830078125,
+      "step": 4000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.19674755381604697,
+      "eval_loss": 5.325084686279297,
+      "eval_runtime": 6.607,
+      "eval_samples_per_second": 151.355,
+      "eval_steps_per_second": 18.919,
+      "step": 4000
     }
   ],
   "logging_steps": 100,