Checkpoint at step 50000

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e852e31fadf2566fdba7ebc8d456578820b724fcc2d3481b296277321c7a528a
+oid sha256:bed344f4e6ec0ab4f3b67a3c9cc2fbf1862edaa4c9779ef7eec194aad98c31ee
 size 828512040

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d795647eb1b830bb555594115e0d0e2dd648b2292077362acfc38684c8d42be
+oid sha256:3f4055448e576b241a37c2bf8f91a66d9e3fd12006d1b596faa06c7c44bd5dd6
 size 1657183499

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24283bdedf9664a219b45889b2cb297750d51485079a99ec4cc794379d885fef
+oid sha256:12ece05b2ede9df156262979286e339c055e8c2fe8447054b30a9471fde11074
 size 14709

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40e39c62dc7958d0b4ec9e70f0a92b9bbef106b669c87a83bd7746958ec30824
+oid sha256:d104a96a640ef80c3c92be9df98931218ac558e8bc30b38f25df2a21f8fb58ea
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4550786e39369e9c3ca0437120bbc321596c5a6d7f9b1cef8d6993ef9b199588
+oid sha256:0ac48ccb0ca9cc2f554de5934a471ee333c535d947c0271e83c358c05bdf5522
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.6176016469377251,
+  "epoch": 0.7720020586721564,
   "eval_steps": 500,
-  "global_step": 40000,
+  "global_step": 50000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -288,6 +288,76 @@
       "learning_rate": 1.3824169716059106e-05,
       "loss": 0.017378036499023437,
       "step": 40000
+    },
+    {
+      "epoch": 0.6330416881111683,
+      "grad_norm": 0.022699156776070595,
+      "learning_rate": 1.3669770098970156e-05,
+      "loss": 0.01683005142211914,
+      "step": 41000
+    },
+    {
+      "epoch": 0.6484817292846115,
+      "grad_norm": 0.15368737280368805,
+      "learning_rate": 1.3515370481881207e-05,
+      "loss": 0.016543420791625976,
+      "step": 42000
+    },
+    {
+      "epoch": 0.6639217704580546,
+      "grad_norm": 0.26471880078315735,
+      "learning_rate": 1.3360970864792256e-05,
+      "loss": 0.01756926727294922,
+      "step": 43000
+    },
+    {
+      "epoch": 0.6793618116314977,
+      "grad_norm": 0.20446893572807312,
+      "learning_rate": 1.3206571247703305e-05,
+      "loss": 0.016958894729614257,
+      "step": 44000
+    },
+    {
+      "epoch": 0.6948018528049408,
+      "grad_norm": 0.1649283766746521,
+      "learning_rate": 1.3052171630614356e-05,
+      "loss": 0.01631136894226074,
+      "step": 45000
+    },
+    {
+      "epoch": 0.7102418939783839,
+      "grad_norm": 0.3699822425842285,
+      "learning_rate": 1.2897772013525407e-05,
+      "loss": 0.01603019714355469,
+      "step": 46000
+    },
+    {
+      "epoch": 0.7256819351518271,
+      "grad_norm": 0.016359159722924232,
+      "learning_rate": 1.2743372396436458e-05,
+      "loss": 0.01701980781555176,
+      "step": 47000
+    },
+    {
+      "epoch": 0.7411219763252702,
+      "grad_norm": 0.35645583271980286,
+      "learning_rate": 1.2588972779347508e-05,
+      "loss": 0.017957780838012697,
+      "step": 48000
+    },
+    {
+      "epoch": 0.7565620174987133,
+      "grad_norm": 0.5092584490776062,
+      "learning_rate": 1.2434573162258559e-05,
+      "loss": 0.016010673522949218,
+      "step": 49000
+    },
+    {
+      "epoch": 0.7720020586721564,
+      "grad_norm": 0.1566636711359024,
+      "learning_rate": 1.228017354516961e-05,
+      "loss": 0.015569474220275879,
+      "step": 50000
     }
   ],
   "logging_steps": 1000,
@@ -307,7 +377,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.183935693796557e+17,
+  "total_flos": 7.730943166037606e+17,
   "train_batch_size": 24,
   "trial_name": null,
   "trial_params": null

run/progress.json

@@ -1,35 +1,35 @@
 {
   "state": "training",
-  "global_step": 40000,
+  "global_step": 50000,
   "max_steps": 129534,
-  "epoch": 0.6176016469377251,
-  "elapsed_seconds": 3863.9128347660007,
-  "steps_per_second": 2.5880501004121648,
-  "optimizer_updates_this_process": 10000,
-  "peak_cuda_GiB": 16.14780044555664,
+  "epoch": 0.7720020586721564,
+  "elapsed_seconds": 7970.391427626004,
+  "steps_per_second": 2.509287050906738,
+  "optimizer_updates_this_process": 20000,
+  "peak_cuda_GiB": 16.147835731506348,
   "world_size": 1,
   "effective_batch_size": 72,
   "recent_metrics": [
     {
-      "loss": 0.017340713500976563,
-      "grad_norm": 0.1389719396829605,
-      "learning_rate": 1.4287368567325955e-05,
-      "epoch": 0.5712815234173958,
-      "step": 37000
+      "loss": 0.01701980781555176,
+      "grad_norm": 0.016359159722924232,
+      "learning_rate": 1.2743372396436458e-05,
+      "epoch": 0.7256819351518271,
+      "step": 47000
     },
     {
-      "loss": 0.016769933700561523,
-      "grad_norm": 0.020788291469216347,
-      "learning_rate": 1.4132968950237004e-05,
-      "epoch": 0.5867215645908389,
-      "step": 38000
+      "loss": 0.017957780838012697,
+      "grad_norm": 0.35645583271980286,
+      "learning_rate": 1.2588972779347508e-05,
+      "epoch": 0.7411219763252702,
+      "step": 48000
     },
     {
-      "loss": 0.01715911293029785,
-      "grad_norm": 0.014325047843158245,
-      "learning_rate": 1.3978569333148055e-05,
-      "epoch": 0.602161605764282,
-      "step": 39000
+      "loss": 0.016010673522949218,
+      "grad_norm": 0.5092584490776062,
+      "learning_rate": 1.2434573162258559e-05,
+      "epoch": 0.7565620174987133,
+      "step": 49000
     }
   ]
 }