Training in progress, step 40000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c650ad8d73e16f826075e38554ceb1b67a79c47e995db17a3f0f293e890e0459
+oid sha256:60cb01bc9e67975e1bb1ed22126a2e438f937a6615e57c3f2220be60495920b2
 size 202193937

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47321182fffa9c08c38cb86c1d3437987336a02a88268582f1c7d4c56bf41c6c
+oid sha256:0fe754e765d5235d6d33bcfe88cdb3a933ccbfc6b7ea5ab60542395d9d85e684
 size 102501541

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c9a68d60eece42765d7548ebff0f60edb5bcb4873ea380215a947751555abf4
+oid sha256:24c2467e9acf06a94a3ca165e312978b81f51dda2e7f32b8607c9d3f773c46e2
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e58a1e19a5ade8deb320d2b0995e1a813a262cbe6344f151bd361d5b954d52d4
-size 14567
+oid sha256:f10426f56c7c01ca78fb77f1b5bd9d6f429ab8feed08a9636c7f90c77533d382
+size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f12cd4c88df48efaa86041a90394bcd19fddebbb2abbf74310d7f26ad92deb28
+oid sha256:50c64be9bba64fb9b54421cbdb5d49bbf62e8dede6378e3090056f6fda12cc3b
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61e25d8922ec72503958c98e324dce1bc02c01c238a4ff522944093bda369731
+oid sha256:15fb56fed836ca09e88803cfd50127dd08d2e98851e7cca02b6a5cfa6f8d7f94
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:351295f8dc223e9573a0b804178e93ff1e6a57eff2a88f2582612cbd70a79601
-size 14567
+oid sha256:7c328439be9bb4cea23920597b019c57083a44b1906a345c7adf7efb277d292f
+size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e37818c7347e85c2960dfac4edd189581f7103968ade461896de2487ebe407a
+oid sha256:bdb4fabc1b2d87119cb5ed67a0ca4df3e9e1c5d3ea269627d2e42cc2977023fd
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:718669ed444618c272fc053a4dbdbdff7633d0333b7d79e5ee880f66e39dcda6
+oid sha256:15fb56fed836ca09e88803cfd50127dd08d2e98851e7cca02b6a5cfa6f8d7f94
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9358c18d93e737c322e3bff2a2eff70043439e104f00fbc6a288e710efdf4ae5
-size 14503
+oid sha256:0ba5ebd9e984982fc400bda128508420216070834c053705f8708435245bde3f
+size 14439

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff9b79eba2b3c3487d57b5f9509b60a9305753a31bdaa8757ad442412114850e
+oid sha256:a74054da3729955ad5eacae83f875e2df84ec7d4ca3ed3437d6f2cf84557171f
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.9191458071632097,
-  "global_step": 30000,
+  "epoch": 1.2255277428842795,
+  "global_step": 40000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -606,11 +606,211 @@
       "eval_samples_per_second": 1974.102,
       "eval_steps_per_second": 31.586,
       "step": 30000
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.0002999040758362914,
+      "loss": 0.4791,
+      "step": 30500
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00029988584471039094,
+      "loss": 0.4776,
+      "step": 31000
+    },
+    {
+      "epoch": 0.95,
+      "eval_loss": 0.8034030795097351,
+      "eval_runtime": 0.4996,
+      "eval_samples_per_second": 2001.487,
+      "eval_steps_per_second": 32.024,
+      "step": 31000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.0002998660291361054,
+      "loss": 0.476,
+      "step": 31500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.0002998446293301349,
+      "loss": 0.4741,
+      "step": 32000
+    },
+    {
+      "epoch": 0.98,
+      "eval_loss": 0.8003847599029541,
+      "eval_runtime": 0.5033,
+      "eval_samples_per_second": 1986.864,
+      "eval_steps_per_second": 31.79,
+      "step": 32000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002998216455265042,
+      "loss": 0.4724,
+      "step": 32500
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00029979707797656046,
+      "loss": 0.4709,
+      "step": 33000
+    },
+    {
+      "epoch": 1.01,
+      "eval_loss": 0.7997156977653503,
+      "eval_runtime": 0.5174,
+      "eval_samples_per_second": 1932.776,
+      "eval_steps_per_second": 30.924,
+      "step": 33000
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00029977092694897053,
+      "loss": 0.4691,
+      "step": 33500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0002997431927297178,
+      "loss": 0.467,
+      "step": 34000
+    },
+    {
+      "epoch": 1.04,
+      "eval_loss": 0.803820013999939,
+      "eval_runtime": 0.5367,
+      "eval_samples_per_second": 1863.102,
+      "eval_steps_per_second": 29.81,
+      "step": 34000
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00029971387562209936,
+      "loss": 0.4653,
+      "step": 34500
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00029968297594672226,
+      "loss": 0.4638,
+      "step": 35000
+    },
+    {
+      "epoch": 1.07,
+      "eval_loss": 0.8011023998260498,
+      "eval_runtime": 0.5429,
+      "eval_samples_per_second": 1842.037,
+      "eval_steps_per_second": 29.473,
+      "step": 35000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0002996504940415005,
+      "loss": 0.462,
+      "step": 35500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00029961643026165096,
+      "loss": 0.4601,
+      "step": 36000
+    },
+    {
+      "epoch": 1.1,
+      "eval_loss": 0.8029471635818481,
+      "eval_runtime": 0.5067,
+      "eval_samples_per_second": 1973.42,
+      "eval_steps_per_second": 31.575,
+      "step": 36000
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00029958078497968973,
+      "loss": 0.4587,
+      "step": 36500
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0002995435585854278,
+      "loss": 0.457,
+      "step": 37000
+    },
+    {
+      "epoch": 1.13,
+      "eval_loss": 0.7997317910194397,
+      "eval_runtime": 0.5216,
+      "eval_samples_per_second": 1917.135,
+      "eval_steps_per_second": 30.674,
+      "step": 37000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0002995047514859671,
+      "loss": 0.4556,
+      "step": 37500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0002994643641056959,
+      "loss": 0.4541,
+      "step": 38000
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.8004159927368164,
+      "eval_runtime": 0.5088,
+      "eval_samples_per_second": 1965.38,
+      "eval_steps_per_second": 31.446,
+      "step": 38000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.000299422396886284,
+      "loss": 0.4524,
+      "step": 38500
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.0002993788502866783,
+      "loss": 0.4514,
+      "step": 39000
+    },
+    {
+      "epoch": 1.19,
+      "eval_loss": 0.797309160232544,
+      "eval_runtime": 0.5211,
+      "eval_samples_per_second": 1918.834,
+      "eval_steps_per_second": 30.701,
+      "step": 39000
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00029933372478309746,
+      "loss": 0.45,
+      "step": 39500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00029928702086902664,
+      "loss": 0.4485,
+      "step": 40000
+    },
+    {
+      "epoch": 1.23,
+      "eval_loss": 0.7876560688018799,
+      "eval_runtime": 0.5381,
+      "eval_samples_per_second": 1858.547,
+      "eval_steps_per_second": 29.737,
+      "step": 40000
     }
   ],
   "max_steps": 500000,
   "num_train_epochs": 16,
-  "total_flos": 9.584614163235e+20,
+  "total_flos": 1.2779405671460241e+21,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47321182fffa9c08c38cb86c1d3437987336a02a88268582f1c7d4c56bf41c6c
+oid sha256:0fe754e765d5235d6d33bcfe88cdb3a933ccbfc6b7ea5ab60542395d9d85e684
 size 102501541