Training in progress, step 40000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15f2dd6e7f5e53889de211105010b6adf4dd822d1c22632581b39aaeed190045
+oid sha256:d70265470022ea1ea19ba2ecc6b2ba2ab39831157147b92926ca5299368d28f1
 size 202193937

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9fd0a688caf940f8820c737d16d6966081330b7042c1771a3baa0d23b03d571
+oid sha256:07e2b246bf7aa13a5e2f82d46ff1f82b0865e3cc8d30ecbee7edaf22d95ad268
 size 102501541

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0018d2eead563c569a412ce1e4b327dc36b583f10e717176ec4f47af700379f7
+oid sha256:26ce7bafde677dd834b0e717580e9362f8261d8f33ccf9c59c48b6a40577d154
 size 14503

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": 0.6691648822269807,
-  "global_step": 30000,
+  "epoch": 0.8922198429693077,
+  "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": 1247.702,
       "eval_steps_per_second": 19.555,
       "step": 30000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0002999040758362914,
+      "loss": 0.3894,
+      "step": 30500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00029988584471039094,
+      "loss": 0.3869,
+      "step": 31000
+    },
+    {
+      "epoch": 0.69,
+      "eval_loss": 0.3689954876899719,
+      "eval_runtime": 1.9129,
+      "eval_samples_per_second": 1200.807,
+      "eval_steps_per_second": 18.82,
+      "step": 31000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0002998660291361054,
+      "loss": 0.3851,
+      "step": 31500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002998446293301349,
+      "loss": 0.3831,
+      "step": 32000
+    },
+    {
+      "epoch": 0.71,
+      "eval_loss": 0.3645932674407959,
+      "eval_runtime": 1.8679,
+      "eval_samples_per_second": 1229.739,
+      "eval_steps_per_second": 19.273,
+      "step": 32000
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0002998216455265042,
+      "loss": 0.3813,
+      "step": 32500
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00029979707797656046,
+      "loss": 0.3794,
+      "step": 33000
+    },
+    {
+      "epoch": 0.74,
+      "eval_loss": 0.35992059111595154,
+      "eval_runtime": 1.8813,
+      "eval_samples_per_second": 1220.942,
+      "eval_steps_per_second": 19.135,
+      "step": 33000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00029977092694897053,
+      "loss": 0.3773,
+      "step": 33500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0002997431927297178,
+      "loss": 0.3756,
+      "step": 34000
+    },
+    {
+      "epoch": 0.76,
+      "eval_loss": 0.3545241057872772,
+      "eval_runtime": 1.8945,
+      "eval_samples_per_second": 1212.449,
+      "eval_steps_per_second": 19.002,
+      "step": 34000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00029971387562209936,
+      "loss": 0.3739,
+      "step": 34500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00029968297594672226,
+      "loss": 0.3723,
+      "step": 35000
+    },
+    {
+      "epoch": 0.78,
+      "eval_loss": 0.3539046347141266,
+      "eval_runtime": 1.8855,
+      "eval_samples_per_second": 1218.214,
+      "eval_steps_per_second": 19.093,
+      "step": 35000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0002996504940415005,
+      "loss": 0.3712,
+      "step": 35500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00029961643026165096,
+      "loss": 0.3693,
+      "step": 36000
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.352187842130661,
+      "eval_runtime": 1.9629,
+      "eval_samples_per_second": 1170.187,
+      "eval_steps_per_second": 18.34,
+      "step": 36000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00029958078497968973,
+      "loss": 0.3681,
+      "step": 36500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0002995435585854278,
+      "loss": 0.3668,
+      "step": 37000
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 0.3482549488544464,
+      "eval_runtime": 1.9671,
+      "eval_samples_per_second": 1167.724,
+      "eval_steps_per_second": 18.301,
+      "step": 37000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002995047514859671,
+      "loss": 0.365,
+      "step": 37500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0002994643641056959,
+      "loss": 0.3635,
+      "step": 38000
+    },
+    {
+      "epoch": 0.85,
+      "eval_loss": 0.3442147970199585,
+      "eval_runtime": 1.9101,
+      "eval_samples_per_second": 1202.583,
+      "eval_steps_per_second": 18.848,
+      "step": 38000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.000299422396886284,
+      "loss": 0.3627,
+      "step": 38500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002993788502866783,
+      "loss": 0.3613,
+      "step": 39000
+    },
+    {
+      "epoch": 0.87,
+      "eval_loss": 0.3413570523262024,
+      "eval_runtime": 1.8505,
+      "eval_samples_per_second": 1241.299,
+      "eval_steps_per_second": 19.454,
+      "step": 39000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.00029933372478309746,
+      "loss": 0.3602,
+      "step": 39500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00029928702086902664,
+      "loss": 0.3589,
+      "step": 40000
+    },
+    {
+      "epoch": 0.89,
+      "eval_loss": 0.3388674557209015,
+      "eval_runtime": 1.8714,
+      "eval_samples_per_second": 1227.432,
+      "eval_steps_per_second": 19.237,
+      "step": 40000
     }
   ],
   "max_steps": 500000,
   "num_train_epochs": 12,
-  "total_flos": 9.584614163235e+20,
+  "total_flos": 1.277948555098e+21,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9fd0a688caf940f8820c737d16d6966081330b7042c1771a3baa0d23b03d571
+oid sha256:07e2b246bf7aa13a5e2f82d46ff1f82b0865e3cc8d30ecbee7edaf22d95ad268
 size 102501541