Training in progress, step 40000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a06ffce25ec404e52367f5cc0b644052826374a7fdd783f9655fcd285df66d6d
+oid sha256:21e0468f83252fbd857b959ddace7a9f33b4b96fb87a6baf73ead5e873aabb0c
 size 893438545

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c329e242cb4a3f96564fe11ec3eccc7274448a9d37d936eb1f9c1b5bcd37ec91
+oid sha256:8bc7e4def826ce0471f39109f9fdb3314c8bc2030700c1f3907e774f1c01f478
 size 449471589

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3050f6cd4e244bcfafaa04a6d1b955b9b5bab22bee349f2d0379131ff0874e3f
+oid sha256:85d1c2df1a9b702c1ab654e9e5e408f92b9742bb1e3ca9287f5761ef0b320b84
 size 15523

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a980a1351c683c05bfb6454932005b70f1b4871d3735c6a11b55d7963483b96
+oid sha256:c80c9fc8381c1a21d068329ef7b88e8e389828d1de4a65d282cc16135a4f7204
 size 559

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbf786375f854cc04cfc07ff0949c0fe8ff8ba67bd06f47c9df99e37ce7c9a9a
+oid sha256:3321618f4e6ed92cd7e76689475cb3f46497297ac6a6a8342c9bad133a001350
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.6125576939824087,
-  "global_step": 30000,
+  "epoch": 3.483410258643212,
+  "global_step": 40000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -414,11 +414,147 @@
       "eval_samples_per_second": 62.361,
       "eval_steps_per_second": 0.985,
       "step": 30000
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 8.150097177421913e-05,
+      "loss": 0.4223,
+      "step": 30500
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 8.091455446965055e-05,
+      "loss": 0.4215,
+      "step": 31000
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 8.032183951740807e-05,
+      "loss": 0.4197,
+      "step": 31500
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 7.972297316103718e-05,
+      "loss": 0.4206,
+      "step": 32000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 7.911931879453039e-05,
+      "loss": 0.4195,
+      "step": 32500
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 7.850860595403895e-05,
+      "loss": 0.4202,
+      "step": 33000
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 7.789218909766887e-05,
+      "loss": 0.4183,
+      "step": 33500
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 7.727022031705582e-05,
+      "loss": 0.418,
+      "step": 34000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 7.664411309317395e-05,
+      "loss": 0.4179,
+      "step": 34500
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 7.601151251252289e-05,
+      "loss": 0.4168,
+      "step": 35000
+    },
+    {
+      "epoch": 3.05,
+      "eval_loss": 0.3975510597229004,
+      "eval_runtime": 48.348,
+      "eval_samples_per_second": 103.417,
+      "eval_steps_per_second": 1.634,
+      "step": 35000
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 7.537382403643505e-05,
+      "loss": 0.4172,
+      "step": 35500
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 7.473120500500002e-05,
+      "loss": 0.4165,
+      "step": 36000
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7.408511341338508e-05,
+      "loss": 0.417,
+      "step": 36500
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 7.343311918291752e-05,
+      "loss": 0.4172,
+      "step": 37000
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 7.277667323698597e-05,
+      "loss": 0.4163,
+      "step": 37500
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 7.211726318817271e-05,
+      "loss": 0.4136,
+      "step": 38000
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 7.145240886444487e-05,
+      "loss": 0.4144,
+      "step": 38500
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 7.078359153588611e-05,
+      "loss": 0.414,
+      "step": 39000
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 7.011097622316453e-05,
+      "loss": 0.4119,
+      "step": 39500
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 6.943472888404397e-05,
+      "loss": 0.413,
+      "step": 40000
+    },
+    {
+      "epoch": 3.48,
+      "eval_loss": 0.39355719089508057,
+      "eval_runtime": 61.3446,
+      "eval_samples_per_second": 81.507,
+      "eval_steps_per_second": 1.288,
+      "step": 40000
     }
   ],
   "max_steps": 100000,
   "num_train_epochs": 9,
-  "total_flos": 1.4128888515849526e+21,
+  "total_flos": 1.8838402100126674e+21,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c329e242cb4a3f96564fe11ec3eccc7274448a9d37d936eb1f9c1b5bcd37ec91
+oid sha256:8bc7e4def826ce0471f39109f9fdb3314c8bc2030700c1f3907e774f1c01f478
 size 449471589