Training in progress, step 4000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f907b0bed5e2f8cf858db047aa22f79320ace855462d3e01515d8384d133c06
+oid sha256:1568527f327df07256f5a6c4e9bbd9b77a59a0494cbe6ba8a95a0e7c9329df65
 size 1976267

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ed1d8ea23bc21d353cc09795bf8da7e15e7897910a93e297620cb29f241e863
+oid sha256:4c813450e0c95fd384d2ff2898b39d648674b117ab52e8fd7daff5eed04513e6
 size 741336195

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1598f69323c5211489e239f2a4a33d79e9afe5826ee72973f1d7e1a5afa7e09a
+oid sha256:fa51b6645b4eccecb101cadb683ef0e972138fa3a9aba5da3ee559d38423cf09
 size 14455

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc933910f13fa41b7f95dbbbc23336663568ef33e75bf367c77b2082bd938180
+oid sha256:3d04230f28149ed35951faa6c0efeb666df4e8906da2d72b8b274de5d8730258
 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": 114.538,
       "eval_steps_per_second": 14.317,
       "step": 2000
+    },
+    {
+      "epoch": 0.042,
+      "grad_norm": 5.402246475219727,
+      "learning_rate": 1.6792e-05,
+      "loss": 6.878953857421875,
+      "step": 2100
+    },
+    {
+      "epoch": 0.044,
+      "grad_norm": 5.448638916015625,
+      "learning_rate": 1.7592000000000004e-05,
+      "loss": 6.885161743164063,
+      "step": 2200
+    },
+    {
+      "epoch": 0.046,
+      "grad_norm": 5.395037651062012,
+      "learning_rate": 1.8392e-05,
+      "loss": 6.817017822265625,
+      "step": 2300
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 7.231200695037842,
+      "learning_rate": 1.9192000000000002e-05,
+      "loss": 6.765516357421875,
+      "step": 2400
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 4.857005596160889,
+      "learning_rate": 1.9992e-05,
+      "loss": 6.7409033203125,
+      "step": 2500
+    },
+    {
+      "epoch": 0.052,
+      "grad_norm": 4.149160861968994,
+      "learning_rate": 1.9999785636239034e-05,
+      "loss": 6.67716064453125,
+      "step": 2600
+    },
+    {
+      "epoch": 0.054,
+      "grad_norm": 3.6910853385925293,
+      "learning_rate": 1.9999133871331222e-05,
+      "loss": 6.673251342773438,
+      "step": 2700
+    },
+    {
+      "epoch": 0.056,
+      "grad_norm": 5.89624547958374,
+      "learning_rate": 1.9998044711915177e-05,
+      "loss": 6.649310302734375,
+      "step": 2800
+    },
+    {
+      "epoch": 0.058,
+      "grad_norm": 5.415604114532471,
+      "learning_rate": 1.999651820563426e-05,
+      "loss": 6.534013061523438,
+      "step": 2900
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 7.183228969573975,
+      "learning_rate": 1.99945544192628e-05,
+      "loss": 6.565379028320312,
+      "step": 3000
+    },
+    {
+      "epoch": 0.06,
+      "eval_accuracy": 0.1556751467710372,
+      "eval_loss": 6.592763900756836,
+      "eval_runtime": 8.7378,
+      "eval_samples_per_second": 114.445,
+      "eval_steps_per_second": 14.306,
+      "step": 3000
+    },
+    {
+      "epoch": 0.062,
+      "grad_norm": 8.597551345825195,
+      "learning_rate": 1.9992153438703174e-05,
+      "loss": 6.53469970703125,
+      "step": 3100
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 3.4981071949005127,
+      "learning_rate": 1.9989315368982053e-05,
+      "loss": 6.47679931640625,
+      "step": 3200
+    },
+    {
+      "epoch": 0.066,
+      "grad_norm": 6.5537614822387695,
+      "learning_rate": 1.99860403342458e-05,
+      "loss": 6.47298828125,
+      "step": 3300
+    },
+    {
+      "epoch": 0.068,
+      "grad_norm": 5.503767967224121,
+      "learning_rate": 1.9982328477755036e-05,
+      "loss": 6.516907348632812,
+      "step": 3400
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 11.562784194946289,
+      "learning_rate": 1.9978179961878405e-05,
+      "loss": 6.511883544921875,
+      "step": 3500
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 36.503265380859375,
+      "learning_rate": 1.9973594968085412e-05,
+      "loss": 6.553932495117188,
+      "step": 3600
+    },
+    {
+      "epoch": 0.074,
+      "grad_norm": 199.27720642089844,
+      "learning_rate": 1.996857369693855e-05,
+      "loss": 7.200830688476563,
+      "step": 3700
+    },
+    {
+      "epoch": 0.076,
+      "grad_norm": 1094.3033447265625,
+      "learning_rate": 1.996311636808449e-05,
+      "loss": 7.753666381835938,
+      "step": 3800
+    },
+    {
+      "epoch": 0.078,
+      "grad_norm": 472.0617370605469,
+      "learning_rate": 1.9957223220244462e-05,
+      "loss": 7.77503173828125,
+      "step": 3900
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 270.61187744140625,
+      "learning_rate": 1.9950894511203855e-05,
+      "loss": 7.76434814453125,
+      "step": 4000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.04172015655577299,
+      "eval_loss": 7.808525085449219,
+      "eval_runtime": 8.5336,
+      "eval_samples_per_second": 117.184,
+      "eval_steps_per_second": 14.648,
+      "step": 4000
     }
   ],
   "logging_steps": 100,