Training in progress, step 18, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b91670d882bddcee3c9d94c67d5ac920520857f25242046f3fa56c85b626af74
+oid sha256:31dde8579a0157dbc47a9b5157de18564af0aad9b898db302c80bc1ecc7e01e0
 size 671466706

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de73841b08413225d23adbfab69581b3a541f7ca099be1c1264c9e04031c25fa
+oid sha256:fa9423a907785e1975e1cacdbbc1addd40ea1e7958d6da6deb0c44d64a8b8946
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c1fa12db6e65a79cc008f6035c457ab68bb8c45c7fe7a2790ea24e528e6b743
+oid sha256:a4b679dfcad285902b67d18379c2cc1ca0aebfc7646cb33fa82a3fb8ed15c820
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.010089332632685234,
+  "epoch": 0.020178665265370467,
   "eval_steps": 9,
-  "global_step": 9,
+  "global_step": 18,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -44,6 +44,35 @@
       "eval_samples_per_second": 8.271,
       "eval_steps_per_second": 4.137,
       "step": 9
+    },
+    {
+      "epoch": 0.013452443510246978,
+      "grad_norm": NaN,
+      "learning_rate": 8.999999999999999e-05,
+      "loss": 0.0,
+      "step": 12
+    },
+    {
+      "epoch": 0.016815554387808723,
+      "grad_norm": NaN,
+      "learning_rate": 0.0001125,
+      "loss": 0.0,
+      "step": 15
+    },
+    {
+      "epoch": 0.020178665265370467,
+      "grad_norm": NaN,
+      "learning_rate": 0.000135,
+      "loss": 0.0,
+      "step": 18
+    },
+    {
+      "epoch": 0.020178665265370467,
+      "eval_loss": NaN,
+      "eval_runtime": 363.475,
+      "eval_samples_per_second": 8.267,
+      "eval_steps_per_second": 4.135,
+      "step": 18
     }
   ],
   "logging_steps": 3,
@@ -63,7 +92,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.818881526883942e+16,
+  "total_flos": 1.1488560963334963e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null