Training in progress, step 40000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c77af21e597e76684f95ce2433c557649a2b2d1153daf583d6791af200d0ac3a
+oid sha256:c55549ffb5230326e84fdfa6241f0ddb469bb20dd272bd9b2106b1eca6646fde
 size 641630264

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8b05322893de9b8a2e8cfba7ec7dae5f3948555097181441494ebf144a22644
+oid sha256:c0e1da9a673e80e984ea3ad3da1f22102f406d5f65e29af84ba25840dfad38ab
 size 1283324282

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e8920ec89da786a5fc2fb4b9f1869b6bd25c311f6d06cbb6e1066128e3d88f7
+oid sha256:30062d054cd8f10917e3362fb81968d198e989a8d8f5cf9834b4735f11025dbb
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ade4cf88cb30f2d9d39e4476504fc3c8f6aaad1f18169e4d13a5a46ada49f04b
+oid sha256:48ec3b25a48ffdccce52dc6d32ace0b24a39b364bc280dcb7caf619a4463915f
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2fb97a8c11271b16693ce01c66b8779cf869bbc6de203b8f3da783171afdf29
+oid sha256:dad8d7bb7d9da51dc65baece0abc070e82a5f833c2b15381f699042e4af78f88
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.7449344457687723,
+  "epoch": 0.9932459276916965,
   "eval_steps": 5000,
-  "global_step": 30000,
+  "global_step": 40000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -476,6 +476,162 @@
       "eval_samples_per_second": 308.364,
       "eval_steps_per_second": 9.636,
       "step": 30000
+    },
+    {
+      "epoch": 0.7573500198649186,
+      "grad_norm": 3.0172479152679443,
+      "learning_rate": 0.0004264099620682771,
+      "loss": 2.4733,
+      "step": 30500
+    },
+    {
+      "epoch": 0.7697655939610648,
+      "grad_norm": 2.6325442790985107,
+      "learning_rate": 0.000425162208025554,
+      "loss": 2.4721,
+      "step": 31000
+    },
+    {
+      "epoch": 0.7821811680572109,
+      "grad_norm": 2.826345682144165,
+      "learning_rate": 0.0004239144539828309,
+      "loss": 2.4692,
+      "step": 31500
+    },
+    {
+      "epoch": 0.7945967421533572,
+      "grad_norm": 2.456289291381836,
+      "learning_rate": 0.00042266919544819325,
+      "loss": 2.4385,
+      "step": 32000
+    },
+    {
+      "epoch": 0.8070123162495034,
+      "grad_norm": 2.4803292751312256,
+      "learning_rate": 0.0004214214414054702,
+      "loss": 2.439,
+      "step": 32500
+    },
+    {
+      "epoch": 0.8194278903456496,
+      "grad_norm": 2.6469247341156006,
+      "learning_rate": 0.00042017618287083254,
+      "loss": 2.4729,
+      "step": 33000
+    },
+    {
+      "epoch": 0.8318434644417958,
+      "grad_norm": 2.7024786472320557,
+      "learning_rate": 0.0004189284288281094,
+      "loss": 2.4244,
+      "step": 33500
+    },
+    {
+      "epoch": 0.844259038537942,
+      "grad_norm": 2.847285270690918,
+      "learning_rate": 0.0004176806747853863,
+      "loss": 2.4636,
+      "step": 34000
+    },
+    {
+      "epoch": 0.8566746126340882,
+      "grad_norm": 2.453200340270996,
+      "learning_rate": 0.0004164329207426632,
+      "loss": 2.4524,
+      "step": 34500
+    },
+    {
+      "epoch": 0.8690901867302344,
+      "grad_norm": 2.49642276763916,
+      "learning_rate": 0.0004151876622080256,
+      "loss": 2.4457,
+      "step": 35000
+    },
+    {
+      "epoch": 0.8690901867302344,
+      "eval_loss": 1.4776599407196045,
+      "eval_runtime": 4154.52,
+      "eval_samples_per_second": 310.19,
+      "eval_steps_per_second": 9.694,
+      "step": 35000
+    },
+    {
+      "epoch": 0.8815057608263807,
+      "grad_norm": 2.576984405517578,
+      "learning_rate": 0.00041393990816530245,
+      "loss": 2.4149,
+      "step": 35500
+    },
+    {
+      "epoch": 0.8939213349225268,
+      "grad_norm": 3.0729165077209473,
+      "learning_rate": 0.0004126921541225794,
+      "loss": 2.4067,
+      "step": 36000
+    },
+    {
+      "epoch": 0.906336909018673,
+      "grad_norm": 2.7619829177856445,
+      "learning_rate": 0.0004114444000798563,
+      "loss": 2.4121,
+      "step": 36500
+    },
+    {
+      "epoch": 0.9187524831148193,
+      "grad_norm": 3.5316452980041504,
+      "learning_rate": 0.0004101991415452186,
+      "loss": 2.3781,
+      "step": 37000
+    },
+    {
+      "epoch": 0.9311680572109654,
+      "grad_norm": 2.7174599170684814,
+      "learning_rate": 0.0004089538830105809,
+      "loss": 2.4013,
+      "step": 37500
+    },
+    {
+      "epoch": 0.9435836313071116,
+      "grad_norm": 13.372625350952148,
+      "learning_rate": 0.00040771611100019967,
+      "loss": 2.5449,
+      "step": 38000
+    },
+    {
+      "epoch": 0.9559992054032579,
+      "grad_norm": 11.173745155334473,
+      "learning_rate": 0.0004064883210221601,
+      "loss": 6.7867,
+      "step": 38500
+    },
+    {
+      "epoch": 0.968414779499404,
+      "grad_norm": 0.9608703255653381,
+      "learning_rate": 0.000405240566979437,
+      "loss": 7.7097,
+      "step": 39000
+    },
+    {
+      "epoch": 0.9808303535955503,
+      "grad_norm": 8286.201171875,
+      "learning_rate": 0.0004039928129367139,
+      "loss": 7.7291,
+      "step": 39500
+    },
+    {
+      "epoch": 0.9932459276916965,
+      "grad_norm": 2.1965973377227783,
+      "learning_rate": 0.00040274505889399085,
+      "loss": 7.7205,
+      "step": 40000
+    },
+    {
+      "epoch": 0.9932459276916965,
+      "eval_loss": 7.322892665863037,
+      "eval_runtime": 4170.5156,
+      "eval_samples_per_second": 309.0,
+      "eval_steps_per_second": 9.656,
+      "step": 40000
     }
   ],
   "logging_steps": 500,
@@ -495,7 +651,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.260592966550446e+17,
+  "total_flos": 1.6807169819837645e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null