Training in progress, step 30000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0da2970bb6a3a42f4eead306605a961254c0210bad87f083e81965051a1ad3a
+oid sha256:c650ad8d73e16f826075e38554ceb1b67a79c47e995db17a3f0f293e890e0459
 size 202193937

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40868d6c39d293c12d7773c27c15f6382f00c3ba8e481d497b2c196ab25b69a3
+oid sha256:47321182fffa9c08c38cb86c1d3437987336a02a88268582f1c7d4c56bf41c6c
 size 102501541

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b33f31ec9e20c699cd93c0e314986de5d82cb7c411b0e3c7eee0d17f6a92609c
+oid sha256:0c9a68d60eece42765d7548ebff0f60edb5bcb4873ea380215a947751555abf4
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:316d605eec494bdea5793151464c1da876fa7ec2313f7cda46a35a7d559d963f
-size 14503
+oid sha256:e58a1e19a5ade8deb320d2b0995e1a813a262cbe6344f151bd361d5b954d52d4
+size 14567

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31fe7f2ccf79f0fbd307b03672872f533b41579145e63a7888aec3a5ade7df01
-size 14439
+oid sha256:f12cd4c88df48efaa86041a90394bcd19fddebbb2abbf74310d7f26ad92deb28
+size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78efd1d120d3afada7840d4e2f7658c3af4d720391b574486645b700463cee2d
+oid sha256:61e25d8922ec72503958c98e324dce1bc02c01c238a4ff522944093bda369731
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e5c86a710cd0b786d4795e0813f5ead76a96818b39198d12f9f42f9e7430fc1
-size 14439
+oid sha256:351295f8dc223e9573a0b804178e93ff1e6a57eff2a88f2582612cbd70a79601
+size 14567

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1056e9825f4d978bcf0e45dd46c3696fe25102ae04b2b7933b12a49c991f391e
-size 14567
+oid sha256:3e37818c7347e85c2960dfac4edd189581f7103968ade461896de2487ebe407a
+size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdf986c9430e05e7ba964c9e0374d3c9a4f51eff82b71408f66ef54212334841
-size 14439
+oid sha256:718669ed444618c272fc053a4dbdbdff7633d0333b7d79e5ee880f66e39dcda6
+size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81cd2c15231d48fe730759995854c7a6c86a749356ccc31327860314d1e21be9
+oid sha256:9358c18d93e737c322e3bff2a2eff70043439e104f00fbc6a288e710efdf4ae5
 size 14503

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c84272d9e8a15cd78c706d16b77833d53dbfbd182e8ad79e3cd658ef6c3eaaf6
+oid sha256:ff9b79eba2b3c3487d57b5f9509b60a9305753a31bdaa8757ad442412114850e
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.6127638714421397,
-  "global_step": 20000,
+  "epoch": 0.9191458071632097,
+  "global_step": 30000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -406,11 +406,211 @@
       "eval_samples_per_second": 2001.549,
       "eval_steps_per_second": 32.025,
       "step": 20000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00024599999999999996,
+      "loss": 0.516,
+      "step": 20500
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00025199999999999995,
+      "loss": 0.5136,
+      "step": 21000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.8466567993164062,
+      "eval_runtime": 0.521,
+      "eval_samples_per_second": 1919.499,
+      "eval_steps_per_second": 30.712,
+      "step": 21000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.000258,
+      "loss": 0.5112,
+      "step": 21500
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00026399999999999997,
+      "loss": 0.5089,
+      "step": 22000
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 0.8465701937675476,
+      "eval_runtime": 0.5015,
+      "eval_samples_per_second": 1994.167,
+      "eval_steps_per_second": 31.907,
+      "step": 22000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00027,
+      "loss": 0.5067,
+      "step": 22500
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.000276,
+      "loss": 0.5048,
+      "step": 23000
+    },
+    {
+      "epoch": 0.7,
+      "eval_loss": 0.8367106914520264,
+      "eval_runtime": 0.5157,
+      "eval_samples_per_second": 1939.06,
+      "eval_steps_per_second": 31.025,
+      "step": 23000
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00028199999999999997,
+      "loss": 0.5029,
+      "step": 23500
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00028799999999999995,
+      "loss": 0.5009,
+      "step": 24000
+    },
+    {
+      "epoch": 0.74,
+      "eval_loss": 0.8282895088195801,
+      "eval_runtime": 0.5191,
+      "eval_samples_per_second": 1926.314,
+      "eval_steps_per_second": 30.821,
+      "step": 24000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.000294,
+      "loss": 0.4986,
+      "step": 24500
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0003,
+      "loss": 0.4968,
+      "step": 25000
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.8393005132675171,
+      "eval_runtime": 0.5106,
+      "eval_samples_per_second": 1958.371,
+      "eval_steps_per_second": 31.334,
+      "step": 25000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00029999920715161553,
+      "loss": 0.495,
+      "step": 25500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0002999968286151326,
+      "loss": 0.4932,
+      "step": 26000
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.8276752829551697,
+      "eval_runtime": 0.5157,
+      "eval_samples_per_second": 1939.179,
+      "eval_steps_per_second": 31.027,
+      "step": 26000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0002999928644165624,
+      "loss": 0.4916,
+      "step": 26500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0002999873145992569,
+      "loss": 0.4899,
+      "step": 27000
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 0.8151644468307495,
+      "eval_runtime": 0.5093,
+      "eval_samples_per_second": 1963.529,
+      "eval_steps_per_second": 31.416,
+      "step": 27000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.000299980179223908,
+      "loss": 0.4883,
+      "step": 27500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002999714583685469,
+      "loss": 0.4868,
+      "step": 28000
+    },
+    {
+      "epoch": 0.86,
+      "eval_loss": 0.8103233575820923,
+      "eval_runtime": 0.5126,
+      "eval_samples_per_second": 1950.786,
+      "eval_steps_per_second": 31.213,
+      "step": 28000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00029996115212854366,
+      "loss": 0.4855,
+      "step": 28500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00029994926061660554,
+      "loss": 0.4841,
+      "step": 29000
+    },
+    {
+      "epoch": 0.89,
+      "eval_loss": 0.8084909319877625,
+      "eval_runtime": 0.5298,
+      "eval_samples_per_second": 1887.571,
+      "eval_steps_per_second": 30.201,
+      "step": 29000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0002999357839627762,
+      "loss": 0.4823,
+      "step": 29500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00029992072231443425,
+      "loss": 0.4806,
+      "step": 30000
+    },
+    {
+      "epoch": 0.92,
+      "eval_loss": 0.8131279945373535,
+      "eval_runtime": 0.5066,
+      "eval_samples_per_second": 1974.102,
+      "eval_steps_per_second": 31.586,
+      "step": 30000
     }
   ],
   "max_steps": 500000,
   "num_train_epochs": 16,
-  "total_flos": 6.38974277549e+20,
+  "total_flos": 9.584614163235e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40868d6c39d293c12d7773c27c15f6382f00c3ba8e481d497b2c196ab25b69a3
+oid sha256:47321182fffa9c08c38cb86c1d3437987336a02a88268582f1c7d4c56bf41c6c
 size 102501541