Training in progress, step 70000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11cf0c7d6b061f7dad52edce9e5088c7de01b0d50fde085c881cac1ce38869f0
-size 50044241
+oid sha256:c390c19518ce2505f6a86ec1e2f47d0d41f2396b4d8d3b3070b634f49cd1065d
+size 50044689

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32de19aae4938ccc6ceff2fe466b7bb7987521081cfbb1f7e1384e10682c2045
+oid sha256:9fec60f3309de39871877c86c47238ea77b026c2a586d001d07ccd4e052fc5ce
 size 25761253

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe81dd181a7243c753033870695d21544789f2b7d0e68df1f7bb5c7aa16b0ba0
+oid sha256:456f6e75773c2103858da7be2712eb3dc4e020101d1bbc59dc5323b907e2cfa7
 size 14503

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a14cae2e3d67695f0ffa4604dbaaa3f7147d70607921bb89c9afa09d57c87a5
+oid sha256:d89de13f8c844c8306762a31aff6dac5ffaadd95c6501762d83dc7939ed9eace
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.6766595289079227,
-  "global_step": 60000,
+  "epoch": 3.122769450392577,
+  "global_step": 70000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1206,11 +1206,211 @@
       "eval_samples_per_second": 1024.23,
       "eval_steps_per_second": 16.052,
       "step": 60000
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0005425079773152721,
+      "loss": 0.5102,
+      "step": 60500
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.0005413455519215879,
+      "loss": 0.5061,
+      "step": 61000
+    },
+    {
+      "epoch": 2.72,
+      "eval_loss": 0.4848201274871826,
+      "eval_runtime": 2.1727,
+      "eval_samples_per_second": 1057.234,
+      "eval_steps_per_second": 16.57,
+      "step": 61000
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 0.0005401727880078093,
+      "loss": 0.502,
+      "step": 61500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0005389897368744289,
+      "loss": 0.4981,
+      "step": 62000
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 0.4745917320251465,
+      "eval_runtime": 2.1553,
+      "eval_samples_per_second": 1065.724,
+      "eval_steps_per_second": 16.703,
+      "step": 62000
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 0.0005377964502719361,
+      "loss": 0.4972,
+      "step": 62500
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.0005365929803985524,
+      "loss": 0.4912,
+      "step": 63000
+    },
+    {
+      "epoch": 2.81,
+      "eval_loss": 0.46808210015296936,
+      "eval_runtime": 2.2693,
+      "eval_samples_per_second": 1012.199,
+      "eval_steps_per_second": 15.864,
+      "step": 63000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.0005353793798979489,
+      "loss": 0.4882,
+      "step": 63500
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.000534155701856943,
+      "loss": 0.4847,
+      "step": 64000
+    },
+    {
+      "epoch": 2.86,
+      "eval_loss": 0.45994389057159424,
+      "eval_runtime": 2.2357,
+      "eval_samples_per_second": 1027.41,
+      "eval_steps_per_second": 16.102,
+      "step": 64000
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 0.0005329219998031763,
+      "loss": 0.4818,
+      "step": 64500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.0005316783277027734,
+      "loss": 0.4792,
+      "step": 65000
+    },
+    {
+      "epoch": 2.9,
+      "eval_loss": 0.4537006914615631,
+      "eval_runtime": 2.1952,
+      "eval_samples_per_second": 1046.372,
+      "eval_steps_per_second": 16.399,
+      "step": 65000
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.0005304247399579808,
+      "loss": 0.4766,
+      "step": 65500
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 0.0005291612914047876,
+      "loss": 0.474,
+      "step": 66000
+    },
+    {
+      "epoch": 2.94,
+      "eval_loss": 0.44910311698913574,
+      "eval_runtime": 2.304,
+      "eval_samples_per_second": 996.96,
+      "eval_steps_per_second": 15.625,
+      "step": 66000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 0.0005278880373105263,
+      "loss": 0.4713,
+      "step": 66500
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0005266050333714561,
+      "loss": 0.4688,
+      "step": 67000
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 0.4437292516231537,
+      "eval_runtime": 2.2404,
+      "eval_samples_per_second": 1025.255,
+      "eval_steps_per_second": 16.068,
+      "step": 67000
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 0.0005253123357103253,
+      "loss": 0.4665,
+      "step": 67500
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 0.0005240100008739177,
+      "loss": 0.464,
+      "step": 68000
+    },
+    {
+      "epoch": 3.03,
+      "eval_loss": 0.4391787052154541,
+      "eval_runtime": 2.1873,
+      "eval_samples_per_second": 1050.17,
+      "eval_steps_per_second": 16.459,
+      "step": 68000
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 0.0005226980858305778,
+      "loss": 0.4616,
+      "step": 68500
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 0.0005213766479677197,
+      "loss": 0.4592,
+      "step": 69000
+    },
+    {
+      "epoch": 3.08,
+      "eval_loss": 0.43239352107048035,
+      "eval_runtime": 2.1819,
+      "eval_samples_per_second": 1052.771,
+      "eval_steps_per_second": 16.5,
+      "step": 69000
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 0.0005200457450893163,
+      "loss": 0.457,
+      "step": 69500
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 0.0005187054354133712,
+      "loss": 0.4547,
+      "step": 70000
+    },
+    {
+      "epoch": 3.12,
+      "eval_loss": 0.4284396469593048,
+      "eval_runtime": 2.1519,
+      "eval_samples_per_second": 1067.405,
+      "eval_steps_per_second": 16.729,
+      "step": 70000
     }
   ],
   "max_steps": 250000,
   "num_train_epochs": 12,
-  "total_flos": 9.609937532304774e+20,
+  "total_flos": 1.1211546241141079e+21,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32de19aae4938ccc6ceff2fe466b7bb7987521081cfbb1f7e1384e10682c2045
+oid sha256:9fec60f3309de39871877c86c47238ea77b026c2a586d001d07ccd4e052fc5ce
 size 25761253