Checkpoint at step 90000

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca6de09e68f27026e91a66dbd57aaed79b5fc05d3c6560ef0ef8a0aa06b67c9f
+oid sha256:302a6d0f14d407ae224c3d2fccb2f02245e6496ecbe942ff11db12a685bd11c5
 size 828512040

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:297adae8c53efe6ce3098b79b1e308d6d2d4567d5b4dd882b8c8e1da25fb7c63
+oid sha256:35924a426e12a685567eb78c829a079b9680c1ab595798d725d77c57b569e6e7
 size 1657183499

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60fc008bd46f7e25637ecedaa7af90e69f669a5c9fd370158e3fe25829f16a4c
+oid sha256:3c8ebedd55fc1128e9c84d0fe4cc89a320875bf6f1300043a918564d69494d62
 size 14709

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee345859562ac25e166f414288a4a0b0d7f0a27bfa731e8b5de92484a0610d91
+oid sha256:a6726b5f2174da6c91c8a0beeb3bcc411f63d2e054b06948878caf572bdd7ef9
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d7f54899ae82e7e73a4bd9537fae3bee65f6692aa03cc3a6f600b0a52dad6a8
+oid sha256:961d06d0987c0d83f5b85948e9a52e93ef0c3ff0a5842c5ffb4cdf6c6ee0f2ba
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.2351981471950593,
+  "epoch": 1.3895985589294906,
   "eval_steps": 500,
-  "global_step": 80000,
+  "global_step": 90000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -568,6 +568,76 @@
       "learning_rate": 7.64818503250112e-06,
       "loss": 0.013527729988098144,
       "step": 80000
+    },
+    {
+      "epoch": 1.2506381883685023,
+      "grad_norm": 0.10128282010555267,
+      "learning_rate": 7.49378541541217e-06,
+      "loss": 0.014240958213806153,
+      "step": 81000
+    },
+    {
+      "epoch": 1.2660782295419455,
+      "grad_norm": 0.027830589562654495,
+      "learning_rate": 7.339385798323221e-06,
+      "loss": 0.012980664253234863,
+      "step": 82000
+    },
+    {
+      "epoch": 1.2815182707153885,
+      "grad_norm": 0.005913850851356983,
+      "learning_rate": 7.184986181234271e-06,
+      "loss": 0.012590609550476074,
+      "step": 83000
+    },
+    {
+      "epoch": 1.2969583118888317,
+      "grad_norm": 0.18359288573265076,
+      "learning_rate": 7.030586564145322e-06,
+      "loss": 0.012307174682617187,
+      "step": 84000
+    },
+    {
+      "epoch": 1.312398353062275,
+      "grad_norm": 0.28034508228302,
+      "learning_rate": 6.876186947056371e-06,
+      "loss": 0.013142762184143066,
+      "step": 85000
+    },
+    {
+      "epoch": 1.327838394235718,
+      "grad_norm": 0.22471600770950317,
+      "learning_rate": 6.721787329967422e-06,
+      "loss": 0.013293880462646485,
+      "step": 86000
+    },
+    {
+      "epoch": 1.3432784354091611,
+      "grad_norm": 0.5216437578201294,
+      "learning_rate": 6.567387712878473e-06,
+      "loss": 0.012979073524475098,
+      "step": 87000
+    },
+    {
+      "epoch": 1.3587184765826041,
+      "grad_norm": 0.10402315109968185,
+      "learning_rate": 6.412988095789523e-06,
+      "loss": 0.012205662727355958,
+      "step": 88000
+    },
+    {
+      "epoch": 1.3741585177560474,
+      "grad_norm": 0.0021522596944123507,
+      "learning_rate": 6.258588478700574e-06,
+      "loss": 0.013783681869506835,
+      "step": 89000
+    },
+    {
+      "epoch": 1.3895985589294906,
+      "grad_norm": 0.12027305364608765,
+      "learning_rate": 6.104188861611624e-06,
+      "loss": 0.012504731178283691,
+      "step": 90000
     }
   ],
   "logging_steps": 1000,
@@ -587,7 +657,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.2376550084179968e+18,
+  "total_flos": 1.3926579605102592e+18,
   "train_batch_size": 24,
   "trial_name": null,
   "trial_params": null

run/progress.json

@@ -1,35 +1,35 @@
 {
   "state": "training",
-  "global_step": 80000,
+  "global_step": 90000,
   "max_steps": 129534,
-  "epoch": 1.2351981471950593,
-  "elapsed_seconds": 20687.525855143,
-  "steps_per_second": 2.416915408353184,
-  "optimizer_updates_this_process": 50000,
+  "epoch": 1.3895985589294906,
+  "elapsed_seconds": 24970.570524402996,
+  "steps_per_second": 2.4028285593780803,
+  "optimizer_updates_this_process": 60000,
   "peak_cuda_GiB": 16.14784526824951,
   "world_size": 1,
   "effective_batch_size": 72,
   "recent_metrics": [
     {
-      "loss": 0.011891658782958985,
-      "grad_norm": 0.015670454129576683,
-      "learning_rate": 8.11138388376797e-06,
-      "epoch": 1.1888780236747298,
-      "step": 77000
+      "loss": 0.012979073524475098,
+      "grad_norm": 0.5216437578201294,
+      "learning_rate": 6.567387712878473e-06,
+      "epoch": 1.3432784354091611,
+      "step": 87000
     },
     {
-      "loss": 0.012898669242858887,
-      "grad_norm": 0.16066931188106537,
-      "learning_rate": 7.95698426667902e-06,
-      "epoch": 1.2043180648481728,
-      "step": 78000
+      "loss": 0.012205662727355958,
+      "grad_norm": 0.10402315109968185,
+      "learning_rate": 6.412988095789523e-06,
+      "epoch": 1.3587184765826041,
+      "step": 88000
     },
     {
-      "loss": 0.013374063491821289,
-      "grad_norm": 0.11956440657377243,
-      "learning_rate": 7.802584649590069e-06,
-      "epoch": 1.219758106021616,
-      "step": 79000
+      "loss": 0.013783681869506835,
+      "grad_norm": 0.0021522596944123507,
+      "learning_rate": 6.258588478700574e-06,
+      "epoch": 1.3741585177560474,
+      "step": 89000
     }
   ]
 }