Checkpoint at step 30000

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94f6ee9e39812a449b20a47601cbe4d8ebad47013f73204ada543575d688d040
+oid sha256:a174ef7ce0c4c91f03dfaa1da446eb56194034d6ebd570bbda05f47837ed2bf0
 size 828512040

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc0e0e9e750051952dd5528fde21d1f3de4d3fb501c156dc8d4ce751d3bfe4d5
+oid sha256:f7427a6a2c676e9c0179d12cf99b17bc116b5ee3931ef4785f3721828477fb9c
 size 1657183499

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c41144581cb3a22c8b3e0a7c459a12a3b8191916fa068bc412870766d538f0f3
+oid sha256:6c03aaf8d6a80448551a2ff6a31325875df2fc4b5630b7ab23b4efffd4ba5bfe
 size 14709

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b872419a9498faae546cbedadda703cdb0d83012e62feafdd633e18d8917f37
+oid sha256:69350fe377f4fbb41ad9123978ca0927051a1fcac35b55eafcfef32a09708d4a
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15251ba63b6eaaa134e022af7fa5f9fe504b46520cfa224fff4a4465950ec8ef
+oid sha256:1a6b200d12a55a703157e4869cf5f39ac2652159ef002f661f44b9e4322421a1
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.30880082346886256,
+  "epoch": 0.46320123520329387,
   "eval_steps": 500,
-  "global_step": 20000,
+  "global_step": 30000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -148,6 +148,76 @@
       "learning_rate": 1.6912162057838096e-05,
       "loss": 0.021963685989379884,
       "step": 20000
+    },
+    {
+      "epoch": 0.3242408646423057,
+      "grad_norm": 0.11887926608324051,
+      "learning_rate": 1.675776244074915e-05,
+      "loss": 0.021782083511352537,
+      "step": 21000
+    },
+    {
+      "epoch": 0.33968090581574883,
+      "grad_norm": 0.0387737937271595,
+      "learning_rate": 1.6603362823660198e-05,
+      "loss": 0.021621700286865234,
+      "step": 22000
+    },
+    {
+      "epoch": 0.35512094698919194,
+      "grad_norm": 0.36461085081100464,
+      "learning_rate": 1.6448963206571248e-05,
+      "loss": 0.02196156883239746,
+      "step": 23000
+    },
+    {
+      "epoch": 0.3705609881626351,
+      "grad_norm": 0.34299856424331665,
+      "learning_rate": 1.62945635894823e-05,
+      "loss": 0.022593551635742188,
+      "step": 24000
+    },
+    {
+      "epoch": 0.3860010293360782,
+      "grad_norm": 0.22921200096607208,
+      "learning_rate": 1.614016397239335e-05,
+      "loss": 0.020230865478515624,
+      "step": 25000
+    },
+    {
+      "epoch": 0.4014410705095214,
+      "grad_norm": 0.23404839634895325,
+      "learning_rate": 1.59857643553044e-05,
+      "loss": 0.01947743034362793,
+      "step": 26000
+    },
+    {
+      "epoch": 0.4168811116829645,
+      "grad_norm": 0.4188960790634155,
+      "learning_rate": 1.583136473821545e-05,
+      "loss": 0.019623750686645507,
+      "step": 27000
+    },
+    {
+      "epoch": 0.4323211528564076,
+      "grad_norm": 0.35333889722824097,
+      "learning_rate": 1.56769651211265e-05,
+      "loss": 0.018160568237304688,
+      "step": 28000
+    },
+    {
+      "epoch": 0.44776119402985076,
+      "grad_norm": 0.02362905442714691,
+      "learning_rate": 1.5522565504037554e-05,
+      "loss": 0.01974146842956543,
+      "step": 29000
+    },
+    {
+      "epoch": 0.46320123520329387,
+      "grad_norm": 0.004046312998980284,
+      "learning_rate": 1.53681658869486e-05,
+      "loss": 0.01891119956970215,
+      "step": 30000
     }
   ],
   "logging_steps": 1000,
@@ -167,7 +237,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.093651853339853e+17,
+  "total_flos": 4.638677453096141e+17,
   "train_batch_size": 24,
   "trial_name": null,
   "trial_params": null

run/progress.json

@@ -1,35 +1,35 @@
 {
   "state": "training",
-  "global_step": 20000,
+  "global_step": 30000,
   "max_steps": 129534,
-  "epoch": 0.30880082346886256,
-  "elapsed_seconds": 7519.569151846998,
-  "steps_per_second": 2.6597268535109477,
-  "optimizer_updates_this_process": 20000,
+  "epoch": 0.46320123520329387,
+  "elapsed_seconds": 11153.100829963005,
+  "steps_per_second": 2.689834912942279,
+  "optimizer_updates_this_process": 30000,
   "peak_cuda_GiB": 16.14784526824951,
   "world_size": 1,
   "effective_batch_size": 72,
   "recent_metrics": [
     {
-      "loss": 0.023529365539550783,
-      "grad_norm": 0.12281997501850128,
-      "learning_rate": 1.7375360909104945e-05,
-      "epoch": 0.2624806999485332,
-      "step": 17000
+      "loss": 0.019623750686645507,
+      "grad_norm": 0.4188960790634155,
+      "learning_rate": 1.583136473821545e-05,
+      "epoch": 0.4168811116829645,
+      "step": 27000
     },
     {
-      "loss": 0.023460807800292968,
-      "grad_norm": 0.33700472116470337,
-      "learning_rate": 1.7220961292015998e-05,
-      "epoch": 0.27792074112197634,
-      "step": 18000
+      "loss": 0.018160568237304688,
+      "grad_norm": 0.35333889722824097,
+      "learning_rate": 1.56769651211265e-05,
+      "epoch": 0.4323211528564076,
+      "step": 28000
     },
     {
-      "loss": 0.022678720474243165,
-      "grad_norm": 0.11011330038309097,
-      "learning_rate": 1.7066561674927047e-05,
-      "epoch": 0.29336078229541945,
-      "step": 19000
+      "loss": 0.01974146842956543,
+      "grad_norm": 0.02362905442714691,
+      "learning_rate": 1.5522565504037554e-05,
+      "epoch": 0.44776119402985076,
+      "step": 29000
     }
   ]
 }