Checkpoint at step 100000

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:302a6d0f14d407ae224c3d2fccb2f02245e6496ecbe942ff11db12a685bd11c5
+oid sha256:56b5a13d4150c701abd51efc77da1a43ce77b5714990023428cac3e49d177a8d
 size 828512040

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35924a426e12a685567eb78c829a079b9680c1ab595798d725d77c57b569e6e7
+oid sha256:1e232163e748616f20b19fc434847c2b32b62c446d4d18b0d14156b66cd51537
 size 1657183499

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c8ebedd55fc1128e9c84d0fe4cc89a320875bf6f1300043a918564d69494d62
+oid sha256:e337cce6ad62f148d14c3fd6cd72b90800c38358b0b6fcdb66fa80f2f80ad00f
 size 14709

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6726b5f2174da6c91c8a0beeb3bcc411f63d2e054b06948878caf572bdd7ef9
+oid sha256:6cf85e7928fef52e8e5f5986b3151d88879d53612aa920bb9a39849c5aeac4ad
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:961d06d0987c0d83f5b85948e9a52e93ef0c3ff0a5842c5ffb4cdf6c6ee0f2ba
+oid sha256:a7ef580f26b1f1872d03b95559ecd67d7596de45b58ca2297a19f1f2c1c3aa61
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.3895985589294906,
+  "epoch": 1.5439989706639219,
   "eval_steps": 500,
-  "global_step": 90000,
+  "global_step": 100000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -638,6 +638,76 @@
       "learning_rate": 6.104188861611624e-06,
       "loss": 0.012504731178283691,
       "step": 90000
+    },
+    {
+      "epoch": 1.4050386001029336,
+      "grad_norm": 0.25139009952545166,
+      "learning_rate": 5.949789244522675e-06,
+      "loss": 0.013058713912963867,
+      "step": 91000
+    },
+    {
+      "epoch": 1.4204786412763768,
+      "grad_norm": 0.004542683716863394,
+      "learning_rate": 5.795389627433724e-06,
+      "loss": 0.012640674591064454,
+      "step": 92000
+    },
+    {
+      "epoch": 1.4359186824498198,
+      "grad_norm": 0.11336570978164673,
+      "learning_rate": 5.6409900103447744e-06,
+      "loss": 0.011859203338623047,
+      "step": 93000
+    },
+    {
+      "epoch": 1.451358723623263,
+      "grad_norm": 0.0319698192179203,
+      "learning_rate": 5.4865903932558254e-06,
+      "loss": 0.011702972412109375,
+      "step": 94000
+    },
+    {
+      "epoch": 1.4667987647967062,
+      "grad_norm": 0.10848435759544373,
+      "learning_rate": 5.332190776166876e-06,
+      "loss": 0.011963166236877442,
+      "step": 95000
+    },
+    {
+      "epoch": 1.4822388059701492,
+      "grad_norm": 0.27238816022872925,
+      "learning_rate": 5.177791159077926e-06,
+      "loss": 0.012217957496643067,
+      "step": 96000
+    },
+    {
+      "epoch": 1.4976788471435925,
+      "grad_norm": 0.01449698768556118,
+      "learning_rate": 5.023391541988977e-06,
+      "loss": 0.012590063095092774,
+      "step": 97000
+    },
+    {
+      "epoch": 1.5131188883170354,
+      "grad_norm": 0.07639008015394211,
+      "learning_rate": 4.868991924900027e-06,
+      "loss": 0.012973206520080566,
+      "step": 98000
+    },
+    {
+      "epoch": 1.5285589294904787,
+      "grad_norm": 0.06926650553941727,
+      "learning_rate": 4.714592307811077e-06,
+      "loss": 0.011427496910095215,
+      "step": 99000
+    },
+    {
+      "epoch": 1.5439989706639219,
+      "grad_norm": 0.08272098749876022,
+      "learning_rate": 4.560192690722127e-06,
+      "loss": 0.011253142356872558,
+      "step": 100000
     }
   ],
   "logging_steps": 1000,
@@ -657,7 +727,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.3926579605102592e+18,
+  "total_flos": 1.5473777471834112e+18,
   "train_batch_size": 24,
   "trial_name": null,
   "trial_params": null

run/progress.json

@@ -1,35 +1,35 @@
 {
   "state": "training",
-  "global_step": 90000,
+  "global_step": 100000,
   "max_steps": 129534,
-  "epoch": 1.3895985589294906,
-  "elapsed_seconds": 24970.570524402996,
-  "steps_per_second": 2.4028285593780803,
-  "optimizer_updates_this_process": 60000,
+  "epoch": 1.5439989706639219,
+  "elapsed_seconds": 29312.662812397,
+  "steps_per_second": 2.3880464374050447,
+  "optimizer_updates_this_process": 70000,
   "peak_cuda_GiB": 16.14784526824951,
   "world_size": 1,
   "effective_batch_size": 72,
   "recent_metrics": [
     {
-      "loss": 0.012979073524475098,
-      "grad_norm": 0.5216437578201294,
-      "learning_rate": 6.567387712878473e-06,
-      "epoch": 1.3432784354091611,
-      "step": 87000
+      "loss": 0.012590063095092774,
+      "grad_norm": 0.01449698768556118,
+      "learning_rate": 5.023391541988977e-06,
+      "epoch": 1.4976788471435925,
+      "step": 97000
     },
     {
-      "loss": 0.012205662727355958,
-      "grad_norm": 0.10402315109968185,
-      "learning_rate": 6.412988095789523e-06,
-      "epoch": 1.3587184765826041,
-      "step": 88000
+      "loss": 0.012973206520080566,
+      "grad_norm": 0.07639008015394211,
+      "learning_rate": 4.868991924900027e-06,
+      "epoch": 1.5131188883170354,
+      "step": 98000
     },
     {
-      "loss": 0.013783681869506835,
-      "grad_norm": 0.0021522596944123507,
-      "learning_rate": 6.258588478700574e-06,
-      "epoch": 1.3741585177560474,
-      "step": 89000
+      "loss": 0.011427496910095215,
+      "grad_norm": 0.06926650553941727,
+      "learning_rate": 4.714592307811077e-06,
+      "epoch": 1.5285589294904787,
+      "step": 99000
     }
   ]
 }