Training in progress, step 22000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:751294f048537038e250758d573a29f1e0658edad8e94d39ab9d4d3a09e6f9e8
+oid sha256:5540b02c0c8b9718f4c8b163b89d2536d0a204a96872cc1813830ea2f5ff2def
 size 529337896

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:003047188ae7aacd3b6801dc800cc4445151ccae190e498568ef9ad5e60978c7
+oid sha256:07da83974549f0bd472e89e0306438f6ba6224e07eeffab08f1bdd4c06240b0c
 size 871247243

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91540c364711a6d4c4520973b3a3f4e9bff7842caef75dc08d20a08085c05012
+oid sha256:67beb949b8d9f7d418f57ffcfa3af010c6a734200eee76f8fffbb6c1af5da698
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc15c4e3a3da92e36573426a7711e53525f3873e45550434cb0365a46f480e71
+oid sha256:76fc781feaf7f256475a659aa4c1161b5c2bf2ab5d085f394f920fa586f35178
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.4,
+  "epoch": 0.44,
   "eval_steps": 1000,
-  "global_step": 20000,
+  "global_step": 22000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1588,6 +1588,164 @@
       "eval_samples_per_second": 78.648,
       "eval_steps_per_second": 0.629,
       "step": 20000
+    },
+    {
+      "epoch": 0.402,
+      "grad_norm": 0.18359500169754028,
+      "learning_rate": 0.0013956906105251406,
+      "loss": 3.4094,
+      "step": 20100
+    },
+    {
+      "epoch": 0.404,
+      "grad_norm": 0.1576140820980072,
+      "learning_rate": 0.0013896079203445497,
+      "loss": 3.3935,
+      "step": 20200
+    },
+    {
+      "epoch": 0.406,
+      "grad_norm": 0.22549428045749664,
+      "learning_rate": 0.0013835081874513679,
+      "loss": 3.4006,
+      "step": 20300
+    },
+    {
+      "epoch": 0.408,
+      "grad_norm": 0.15331390500068665,
+      "learning_rate": 0.001377391678667673,
+      "loss": 3.4129,
+      "step": 20400
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.16119052469730377,
+      "learning_rate": 0.0013712586615493736,
+      "loss": 3.4036,
+      "step": 20500
+    },
+    {
+      "epoch": 0.412,
+      "grad_norm": 0.15083439648151398,
+      "learning_rate": 0.0013651094043745067,
+      "loss": 3.3857,
+      "step": 20600
+    },
+    {
+      "epoch": 0.414,
+      "grad_norm": 0.14811307191848755,
+      "learning_rate": 0.0013589441761315021,
+      "loss": 3.4058,
+      "step": 20700
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 0.15205927193164825,
+      "learning_rate": 0.0013527632465074157,
+      "loss": 3.4109,
+      "step": 20800
+    },
+    {
+      "epoch": 0.418,
+      "grad_norm": 0.164772629737854,
+      "learning_rate": 0.0013465668858761326,
+      "loss": 3.3811,
+      "step": 20900
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.15865328907966614,
+      "learning_rate": 0.00134035536528654,
+      "loss": 3.3878,
+      "step": 21000
+    },
+    {
+      "epoch": 0.42,
+      "eval_accuracy": 0.37497260273972605,
+      "eval_loss": 3.3792359828948975,
+      "eval_runtime": 12.1681,
+      "eval_samples_per_second": 82.182,
+      "eval_steps_per_second": 0.657,
+      "step": 21000
+    },
+    {
+      "epoch": 0.422,
+      "grad_norm": 0.18703462183475494,
+      "learning_rate": 0.0013341289564506717,
+      "loss": 3.401,
+      "step": 21100
+    },
+    {
+      "epoch": 0.424,
+      "grad_norm": 0.15354931354522705,
+      "learning_rate": 0.0013278879317318202,
+      "loss": 3.3947,
+      "step": 21200
+    },
+    {
+      "epoch": 0.426,
+      "grad_norm": 0.14713342487812042,
+      "learning_rate": 0.0013216325641326255,
+      "loss": 3.3774,
+      "step": 21300
+    },
+    {
+      "epoch": 0.428,
+      "grad_norm": 0.17548827826976776,
+      "learning_rate": 0.0013153631272831304,
+      "loss": 3.3794,
+      "step": 21400
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.1897333562374115,
+      "learning_rate": 0.001309079895428813,
+      "loss": 3.3955,
+      "step": 21500
+    },
+    {
+      "epoch": 0.432,
+      "grad_norm": 0.1576170176267624,
+      "learning_rate": 0.0013027831434185898,
+      "loss": 3.3884,
+      "step": 21600
+    },
+    {
+      "epoch": 0.434,
+      "grad_norm": 0.18555521965026855,
+      "learning_rate": 0.0012964731466927916,
+      "loss": 3.3695,
+      "step": 21700
+    },
+    {
+      "epoch": 0.436,
+      "grad_norm": 0.15987864136695862,
+      "learning_rate": 0.0012901501812711176,
+      "loss": 3.3891,
+      "step": 21800
+    },
+    {
+      "epoch": 0.438,
+      "grad_norm": 0.16719584167003632,
+      "learning_rate": 0.0012838145237405588,
+      "loss": 3.4056,
+      "step": 21900
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.19731645286083221,
+      "learning_rate": 0.0012774664512432996,
+      "loss": 3.3772,
+      "step": 22000
+    },
+    {
+      "epoch": 0.44,
+      "eval_accuracy": 0.3771624266144814,
+      "eval_loss": 3.3640236854553223,
+      "eval_runtime": 12.4106,
+      "eval_samples_per_second": 80.576,
+      "eval_steps_per_second": 0.645,
+      "step": 22000
     }
   ],
   "logging_steps": 100,