Training in progress, step 75, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a38e22abc216800c8e88eeef8ea47987853da7e184938f098df5541c87ef70aa
+oid sha256:d0ad9b7d7e1358fa7679dc7824553aa048ca5858b093691785077f1f080fd8b4
 size 239452242

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98f17a41a58c437540d59a325163d8438ba1d9ddee00bc7ae18f589070b0100d
+oid sha256:51810a7fc80db2e52c6d1ec91ce272e9e98ced1530a7a9a3d8c291d62faf2f76
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d8615f1aeccd0f9873fc0cf2c0322fde20ac202421c015654a471730a5fa755
+oid sha256:feed204db996a9f7ac3167cc4686c04dccd6605b9b2a7d539da6ff57c605983d
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.33003300330033003,
+  "epoch": 0.49504950495049505,
   "eval_steps": 9,
-  "global_step": 50,
+  "global_step": 75,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -405,6 +405,205 @@
       "learning_rate": 1.1736481776669307e-05,
       "loss": 0.0,
       "step": 50
+    },
+    {
+      "epoch": 0.33663366336633666,
+      "grad_norm": NaN,
+      "learning_rate": 1.1391731009600655e-05,
+      "loss": 0.0,
+      "step": 51
+    },
+    {
+      "epoch": 0.3432343234323432,
+      "grad_norm": NaN,
+      "learning_rate": 1.1045284632676535e-05,
+      "loss": 0.0,
+      "step": 52
+    },
+    {
+      "epoch": 0.34983498349834985,
+      "grad_norm": NaN,
+      "learning_rate": 1.0697564737441254e-05,
+      "loss": 0.0,
+      "step": 53
+    },
+    {
+      "epoch": 0.3564356435643564,
+      "grad_norm": NaN,
+      "learning_rate": 1.0348994967025012e-05,
+      "loss": 0.0,
+      "step": 54
+    },
+    {
+      "epoch": 0.3564356435643564,
+      "eval_loss": NaN,
+      "eval_runtime": 9.4104,
+      "eval_samples_per_second": 13.602,
+      "eval_steps_per_second": 1.7,
+      "step": 54
+    },
+    {
+      "epoch": 0.36303630363036304,
+      "grad_norm": NaN,
+      "learning_rate": 1e-05,
+      "loss": 0.0,
+      "step": 55
+    },
+    {
+      "epoch": 0.3696369636963696,
+      "grad_norm": NaN,
+      "learning_rate": 9.651005032974994e-06,
+      "loss": 0.0,
+      "step": 56
+    },
+    {
+      "epoch": 0.37623762376237624,
+      "grad_norm": NaN,
+      "learning_rate": 9.302435262558748e-06,
+      "loss": 0.0,
+      "step": 57
+    },
+    {
+      "epoch": 0.38283828382838286,
+      "grad_norm": NaN,
+      "learning_rate": 8.954715367323468e-06,
+      "loss": 0.0,
+      "step": 58
+    },
+    {
+      "epoch": 0.38943894389438943,
+      "grad_norm": NaN,
+      "learning_rate": 8.60826899039935e-06,
+      "loss": 0.0,
+      "step": 59
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": NaN,
+      "learning_rate": 8.263518223330698e-06,
+      "loss": 0.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.40264026402640263,
+      "grad_norm": NaN,
+      "learning_rate": 7.92088309182241e-06,
+      "loss": 0.0,
+      "step": 61
+    },
+    {
+      "epoch": 0.40924092409240925,
+      "grad_norm": NaN,
+      "learning_rate": 7.580781044003324e-06,
+      "loss": 0.0,
+      "step": 62
+    },
+    {
+      "epoch": 0.4158415841584158,
+      "grad_norm": NaN,
+      "learning_rate": 7.243626441830009e-06,
+      "loss": 0.0,
+      "step": 63
+    },
+    {
+      "epoch": 0.4158415841584158,
+      "eval_loss": NaN,
+      "eval_runtime": 9.4017,
+      "eval_samples_per_second": 13.615,
+      "eval_steps_per_second": 1.702,
+      "step": 63
+    },
+    {
+      "epoch": 0.42244224422442245,
+      "grad_norm": NaN,
+      "learning_rate": 6.909830056250527e-06,
+      "loss": 0.0,
+      "step": 64
+    },
+    {
+      "epoch": 0.429042904290429,
+      "grad_norm": NaN,
+      "learning_rate": 6.579798566743314e-06,
+      "loss": 0.0,
+      "step": 65
+    },
+    {
+      "epoch": 0.43564356435643564,
+      "grad_norm": NaN,
+      "learning_rate": 6.25393406584088e-06,
+      "loss": 0.0,
+      "step": 66
+    },
+    {
+      "epoch": 0.44224422442244227,
+      "grad_norm": NaN,
+      "learning_rate": 5.932633569242e-06,
+      "loss": 0.0,
+      "step": 67
+    },
+    {
+      "epoch": 0.44884488448844884,
+      "grad_norm": NaN,
+      "learning_rate": 5.616288532109225e-06,
+      "loss": 0.0,
+      "step": 68
+    },
+    {
+      "epoch": 0.45544554455445546,
+      "grad_norm": NaN,
+      "learning_rate": 5.305284372141095e-06,
+      "loss": 0.0,
+      "step": 69
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": NaN,
+      "learning_rate": 5.000000000000003e-06,
+      "loss": 0.0,
+      "step": 70
+    },
+    {
+      "epoch": 0.46864686468646866,
+      "grad_norm": NaN,
+      "learning_rate": 4.700807357667953e-06,
+      "loss": 0.0,
+      "step": 71
+    },
+    {
+      "epoch": 0.4752475247524752,
+      "grad_norm": NaN,
+      "learning_rate": 4.408070965292534e-06,
+      "loss": 0.0,
+      "step": 72
+    },
+    {
+      "epoch": 0.4752475247524752,
+      "eval_loss": NaN,
+      "eval_runtime": 9.4077,
+      "eval_samples_per_second": 13.606,
+      "eval_steps_per_second": 1.701,
+      "step": 72
+    },
+    {
+      "epoch": 0.48184818481848185,
+      "grad_norm": NaN,
+      "learning_rate": 4.12214747707527e-06,
+      "loss": 0.0,
+      "step": 73
+    },
+    {
+      "epoch": 0.4884488448844885,
+      "grad_norm": NaN,
+      "learning_rate": 3.8433852467434175e-06,
+      "loss": 0.0,
+      "step": 74
+    },
+    {
+      "epoch": 0.49504950495049505,
+      "grad_norm": NaN,
+      "learning_rate": 3.5721239031346067e-06,
+      "loss": 0.0,
+      "step": 75
     }
   ],
   "logging_steps": 1,
@@ -424,7 +623,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.8444107907072e+16,
+  "total_flos": 2.803504401874944e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null