Training in progress, step 3500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51f6e02c9049d9eb2741a87e15c4995bca735aa623c1eacd8c972af7d785b970
+oid sha256:6f7ca77c9802a04580d4e31461c12e17e933a3887576c9945c8e484919c0199d
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b0fe8f85e6ddb45c62b62781d7c657ebd54e09d3ba4af58ee61f07f59f2d325
+oid sha256:0af00b7662d798d9ddc39cfcfb857f08567041c5b1d85350af7dc033ca6e0561
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:373099bcd0d3ee5ceb0d1d94a6fdc048cf24515e4576a9744fed78c8d5e543ab
+oid sha256:3bf2e8e416936cf8f439e26111c4885ea7eb78961a1138cdb6b5117961ec9a25
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f37eacb952a10eea80d10445d29e75511960fff96b96070d64acac281d799da8
+oid sha256:540fdd37fedd646a830d2028d95a2e3f201d23e9e3ef83da82d08e9580c78331
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3953019976615906,
-  "best_model_checkpoint": "mikhail_panzo/ceb_b64_le4_s4000/checkpoint-2500",
-  "epoch": 118.81188118811882,
+  "best_metric": 0.39516589045524597,
+  "best_model_checkpoint": "mikhail_panzo/ceb_b64_le4_s4000/checkpoint-3500",
+  "epoch": 138.6138613861386,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -475,6 +475,84 @@
       "eval_samples_per_second": 25.553,
       "eval_steps_per_second": 3.265,
       "step": 3000
+    },
+    {
+      "epoch": 120.79207920792079,
+      "grad_norm": 0.7711001634597778,
+      "learning_rate": 4.755e-05,
+      "loss": 0.3657,
+      "step": 3050
+    },
+    {
+      "epoch": 122.77227722772277,
+      "grad_norm": 0.9125708937644958,
+      "learning_rate": 4.5050000000000004e-05,
+      "loss": 0.3665,
+      "step": 3100
+    },
+    {
+      "epoch": 124.75247524752476,
+      "grad_norm": 1.138588786125183,
+      "learning_rate": 4.2550000000000004e-05,
+      "loss": 0.3714,
+      "step": 3150
+    },
+    {
+      "epoch": 126.73267326732673,
+      "grad_norm": 0.8590717315673828,
+      "learning_rate": 4.0050000000000004e-05,
+      "loss": 0.3696,
+      "step": 3200
+    },
+    {
+      "epoch": 128.7128712871287,
+      "grad_norm": 0.8743947148323059,
+      "learning_rate": 3.7550000000000005e-05,
+      "loss": 0.3652,
+      "step": 3250
+    },
+    {
+      "epoch": 130.69306930693068,
+      "grad_norm": 0.8117642998695374,
+      "learning_rate": 3.505e-05,
+      "loss": 0.3643,
+      "step": 3300
+    },
+    {
+      "epoch": 132.67326732673268,
+      "grad_norm": 0.9085512161254883,
+      "learning_rate": 3.2550000000000005e-05,
+      "loss": 0.3637,
+      "step": 3350
+    },
+    {
+      "epoch": 134.65346534653466,
+      "grad_norm": 0.8282843232154846,
+      "learning_rate": 3.0050000000000002e-05,
+      "loss": 0.3594,
+      "step": 3400
+    },
+    {
+      "epoch": 136.63366336633663,
+      "grad_norm": 0.6631731390953064,
+      "learning_rate": 2.7550000000000002e-05,
+      "loss": 0.3621,
+      "step": 3450
+    },
+    {
+      "epoch": 138.6138613861386,
+      "grad_norm": 0.8194773197174072,
+      "learning_rate": 2.5050000000000002e-05,
+      "loss": 0.3628,
+      "step": 3500
+    },
+    {
+      "epoch": 138.6138613861386,
+      "eval_loss": 0.39516589045524597,
+      "eval_runtime": 7.3015,
+      "eval_samples_per_second": 24.652,
+      "eval_steps_per_second": 3.15,
+      "step": 3500
     }
   ],
   "logging_steps": 50,
@@ -494,7 +572,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.2419037301352416e+16,
+  "total_flos": 3.78207684271391e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null