Training in progress, step 4000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f7ca77c9802a04580d4e31461c12e17e933a3887576c9945c8e484919c0199d
+oid sha256:e35f938dafa0b75d8c4ca77c92d79b765fb12686375a263279c54f290b522762
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0af00b7662d798d9ddc39cfcfb857f08567041c5b1d85350af7dc033ca6e0561
+oid sha256:c4183e38736eff413cb2a7512c613a0d10aa6b21b48c377cdbc1311e4972fd76
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bf2e8e416936cf8f439e26111c4885ea7eb78961a1138cdb6b5117961ec9a25
+oid sha256:546eb67e9ac703715999e4a48906215e9d1d8502c3d871bfb0192cba37d95e1c
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:540fdd37fedd646a830d2028d95a2e3f201d23e9e3ef83da82d08e9580c78331
+oid sha256:4bfcce18e5763d964c4f2e7d8e731979c9aad30909fed83aa5130e2b917f0579
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.39516589045524597,
-  "best_model_checkpoint": "mikhail_panzo/ceb_b64_le4_s4000/checkpoint-3500",
-  "epoch": 138.6138613861386,
+  "best_metric": 0.3941996693611145,
+  "best_model_checkpoint": "mikhail_panzo/ceb_b64_le4_s4000/checkpoint-4000",
+  "epoch": 158.41584158415841,
   "eval_steps": 500,
-  "global_step": 3500,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -553,6 +553,84 @@
       "eval_samples_per_second": 24.652,
       "eval_steps_per_second": 3.15,
       "step": 3500
+    },
+    {
+      "epoch": 140.59405940594058,
+      "grad_norm": 0.9707288146018982,
+      "learning_rate": 2.2550000000000003e-05,
+      "loss": 0.3643,
+      "step": 3550
+    },
+    {
+      "epoch": 142.5742574257426,
+      "grad_norm": 0.7684584856033325,
+      "learning_rate": 2.0050000000000003e-05,
+      "loss": 0.3597,
+      "step": 3600
+    },
+    {
+      "epoch": 144.55445544554456,
+      "grad_norm": 0.759428083896637,
+      "learning_rate": 1.755e-05,
+      "loss": 0.3576,
+      "step": 3650
+    },
+    {
+      "epoch": 146.53465346534654,
+      "grad_norm": 0.6662328839302063,
+      "learning_rate": 1.505e-05,
+      "loss": 0.3559,
+      "step": 3700
+    },
+    {
+      "epoch": 148.5148514851485,
+      "grad_norm": 0.5537445545196533,
+      "learning_rate": 1.255e-05,
+      "loss": 0.3563,
+      "step": 3750
+    },
+    {
+      "epoch": 150.4950495049505,
+      "grad_norm": 0.5476033687591553,
+      "learning_rate": 1.005e-05,
+      "loss": 0.3562,
+      "step": 3800
+    },
+    {
+      "epoch": 152.47524752475246,
+      "grad_norm": 1.056248426437378,
+      "learning_rate": 7.55e-06,
+      "loss": 0.3516,
+      "step": 3850
+    },
+    {
+      "epoch": 154.45544554455446,
+      "grad_norm": 0.556987464427948,
+      "learning_rate": 5.050000000000001e-06,
+      "loss": 0.3575,
+      "step": 3900
+    },
+    {
+      "epoch": 156.43564356435644,
+      "grad_norm": 0.5589704513549805,
+      "learning_rate": 2.55e-06,
+      "loss": 0.3613,
+      "step": 3950
+    },
+    {
+      "epoch": 158.41584158415841,
+      "grad_norm": 0.6661315560340881,
+      "learning_rate": 5.0000000000000004e-08,
+      "loss": 0.3533,
+      "step": 4000
+    },
+    {
+      "epoch": 158.41584158415841,
+      "eval_loss": 0.3941996693611145,
+      "eval_runtime": 7.5266,
+      "eval_samples_per_second": 23.915,
+      "eval_steps_per_second": 3.056,
+      "step": 4000
     }
   ],
   "logging_steps": 50,
@@ -567,12 +645,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 3.78207684271391e+16,
+  "total_flos": 4.322114324975938e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null