Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7965023fb90058fbcca08a10a8e9e39d8cf72c5b9f3d0eb2b25bc715dfb7185
+oid sha256:3ab89fbcd5cd837ff38661f8e6562d1e556debad1cf94429f314679699e56c75
 size 893438545

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5abd226a9e0cc358d3b07162345d74dc2fab3322999fae6a6de0ab8221263680
+oid sha256:650de0629283aadbd2ed9c9c4e4fa6910cc48450c6a7d6491a018378e49a8f08
 size 449471589

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8217c6a790474956dac035c8fd3d42f3d276f02a592aa421e5eed682cc08b9be
+oid sha256:5075119d983cedc61f9e794107c876257cfc00cbcd125777dcdfcea5b38678a0
 size 15523

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46171f332d1e5155156d4999fec5406fc9f03857ff686d0152b9c4ca38519697
+oid sha256:f62cdfaeb53c638813006cc5b9f0699031a6115d20311906da58b054af32fe51
 size 559

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:611657e4e82b38f099b149471e6aaaa5a577926bec8385cc0f6dd743ea4cfd9b
+oid sha256:4fb733c4a27965716bd87114bf2e4699ec7082883b2e3998dcc83abb3d40f744
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.870852564660803,
-  "global_step": 10000,
+  "epoch": 1.741705129321606,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -142,11 +142,147 @@
       "eval_samples_per_second": 67.475,
       "eval_steps_per_second": 1.066,
       "step": 10000
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.833495446081683e-05,
+      "loss": 0.4477,
+      "step": 10500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.809875714179663e-05,
+      "loss": 0.4472,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.785202223022593e-05,
+      "loss": 0.4474,
+      "step": 11500
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.759481060425221e-05,
+      "loss": 0.4452,
+      "step": 12000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.732718572699541e-05,
+      "loss": 0.4443,
+      "step": 12500
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.704921363088947e-05,
+      "loss": 0.4439,
+      "step": 13000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.676096290138969e-05,
+      "loss": 0.4432,
+      "step": 13500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.646311171510121e-05,
+      "loss": 0.4444,
+      "step": 14000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.615453979461317e-05,
+      "loss": 0.443,
+      "step": 14500
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.583590998811311e-05,
+      "loss": 0.4414,
+      "step": 15000
+    },
+    {
+      "epoch": 1.31,
+      "eval_loss": 0.4205426871776581,
+      "eval_runtime": 80.228,
+      "eval_samples_per_second": 62.322,
+      "eval_steps_per_second": 0.985,
+      "step": 15000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.550796803671621e-05,
+      "loss": 0.44,
+      "step": 15500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.516948048567349e-05,
+      "loss": 0.44,
+      "step": 16000
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.482117809726346e-05,
+      "loss": 0.4386,
+      "step": 16500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.446314680988872e-05,
+      "loss": 0.438,
+      "step": 17000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.409547496241208e-05,
+      "loss": 0.4381,
+      "step": 17500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.371901718501017e-05,
+      "loss": 0.4364,
+      "step": 18000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.333235754528592e-05,
+      "loss": 0.4352,
+      "step": 18500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.293633635079047e-05,
+      "loss": 0.436,
+      "step": 19000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.253105131382774e-05,
+      "loss": 0.4357,
+      "step": 19500
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.211660243241619e-05,
+      "loss": 0.4339,
+      "step": 20000
+    },
+    {
+      "epoch": 1.74,
+      "eval_loss": 0.413540244102478,
+      "eval_runtime": 78.0207,
+      "eval_samples_per_second": 64.086,
+      "eval_steps_per_second": 1.013,
+      "step": 20000
     }
   ],
   "max_steps": 100000,
   "num_train_epochs": 9,
-  "total_flos": 4.709861347295232e+20,
+  "total_flos": 9.419374931572379e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5abd226a9e0cc358d3b07162345d74dc2fab3322999fae6a6de0ab8221263680
+oid sha256:650de0629283aadbd2ed9c9c4e4fa6910cc48450c6a7d6491a018378e49a8f08
 size 449471589