Training in progress, step 20000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a028a1582a4cdcab16eaa810c62dd5a92cfee0e12fcbc95547030d75ad6054d
+oid sha256:751294f048537038e250758d573a29f1e0658edad8e94d39ab9d4d3a09e6f9e8
 size 529337896

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b858949970024f5c82faf265e8d537dcfb88c2afcfff507a59916042fcf45114
+oid sha256:003047188ae7aacd3b6801dc800cc4445151ccae190e498568ef9ad5e60978c7
 size 871247243

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68fb3c6eb1526b3dbce1657a61fe1a7b19cc45a43a389d7261b3c6e19d29ffb1
+oid sha256:91540c364711a6d4c4520973b3a3f4e9bff7842caef75dc08d20a08085c05012
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1ec2a8f346b82ffd82d9deb224e5a88186c4ed8b6b71f4b78140ea3176eb4f4
+oid sha256:fc15c4e3a3da92e36573426a7711e53525f3873e45550434cb0365a46f480e71
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.36,
+  "epoch": 0.4,
   "eval_steps": 1000,
-  "global_step": 18000,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1430,6 +1430,164 @@
       "eval_samples_per_second": 78.744,
       "eval_steps_per_second": 0.63,
       "step": 18000
+    },
+    {
+      "epoch": 0.362,
+      "grad_norm": 0.15682971477508545,
+      "learning_rate": 0.0015133615281026555,
+      "loss": 3.426,
+      "step": 18100
+    },
+    {
+      "epoch": 0.364,
+      "grad_norm": 0.15073339641094208,
+      "learning_rate": 0.0015076744925048868,
+      "loss": 3.4358,
+      "step": 18200
+    },
+    {
+      "epoch": 0.366,
+      "grad_norm": 0.18040859699249268,
+      "learning_rate": 0.0015019652495801786,
+      "loss": 3.4261,
+      "step": 18300
+    },
+    {
+      "epoch": 0.368,
+      "grad_norm": 0.18911828100681305,
+      "learning_rate": 0.0014962340490693121,
+      "loss": 3.4359,
+      "step": 18400
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.15816016495227814,
+      "learning_rate": 0.0014904811416735636,
+      "loss": 3.4216,
+      "step": 18500
+    },
+    {
+      "epoch": 0.372,
+      "grad_norm": 0.17654721438884735,
+      "learning_rate": 0.0014847067790437396,
+      "loss": 3.4347,
+      "step": 18600
+    },
+    {
+      "epoch": 0.374,
+      "grad_norm": 0.17527590692043304,
+      "learning_rate": 0.0014789112137691682,
+      "loss": 3.4256,
+      "step": 18700
+    },
+    {
+      "epoch": 0.376,
+      "grad_norm": 0.15412954986095428,
+      "learning_rate": 0.001473094699366649,
+      "loss": 3.4074,
+      "step": 18800
+    },
+    {
+      "epoch": 0.378,
+      "grad_norm": 0.1891617327928543,
+      "learning_rate": 0.0014672574902693661,
+      "loss": 3.4338,
+      "step": 18900
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.16060151159763336,
+      "learning_rate": 0.001461399841815754,
+      "loss": 3.4372,
+      "step": 19000
+    },
+    {
+      "epoch": 0.38,
+      "eval_accuracy": 0.3734951076320939,
+      "eval_loss": 3.3982865810394287,
+      "eval_runtime": 12.1716,
+      "eval_samples_per_second": 82.158,
+      "eval_steps_per_second": 0.657,
+      "step": 19000
+    },
+    {
+      "epoch": 0.382,
+      "grad_norm": 0.17226573824882507,
+      "learning_rate": 0.0014555220102383326,
+      "loss": 3.4123,
+      "step": 19100
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 0.15569022297859192,
+      "learning_rate": 0.0014496242526524964,
+      "loss": 3.4226,
+      "step": 19200
+    },
+    {
+      "epoch": 0.386,
+      "grad_norm": 0.16819970309734344,
+      "learning_rate": 0.001443706827045269,
+      "loss": 3.4294,
+      "step": 19300
+    },
+    {
+      "epoch": 0.388,
+      "grad_norm": 0.14779409766197205,
+      "learning_rate": 0.0014377699922640157,
+      "loss": 3.4233,
+      "step": 19400
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.1541208177804947,
+      "learning_rate": 0.0014318140080051228,
+      "loss": 3.4087,
+      "step": 19500
+    },
+    {
+      "epoch": 0.392,
+      "grad_norm": 0.1646755486726761,
+      "learning_rate": 0.0014258391348026362,
+      "loss": 3.398,
+      "step": 19600
+    },
+    {
+      "epoch": 0.394,
+      "grad_norm": 0.18935076892375946,
+      "learning_rate": 0.0014198456340168662,
+      "loss": 3.4231,
+      "step": 19700
+    },
+    {
+      "epoch": 0.396,
+      "grad_norm": 0.1867913156747818,
+      "learning_rate": 0.0014138337678229528,
+      "loss": 3.4129,
+      "step": 19800
+    },
+    {
+      "epoch": 0.398,
+      "grad_norm": 0.15837714076042175,
+      "learning_rate": 0.0014078037991993992,
+      "loss": 3.3952,
+      "step": 19900
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.15315821766853333,
+      "learning_rate": 0.0014017559919165675,
+      "loss": 3.4155,
+      "step": 20000
+    },
+    {
+      "epoch": 0.4,
+      "eval_accuracy": 0.37439138943248534,
+      "eval_loss": 3.387746572494507,
+      "eval_runtime": 12.7148,
+      "eval_samples_per_second": 78.648,
+      "eval_steps_per_second": 0.629,
+      "step": 20000
     }
   ],
   "logging_steps": 100,