Training in progress, step 3500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a736e732330ece0b457cbaf5c5b4351b5d3db83c02791cf60a402f193e6be32
+oid sha256:e88bcddd088e858a9bb66970598a4a31f3a99fa875613237e38373a495920782
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:621173d0409fc5035f339553890479dbf13f20436f2751f19fa0d3f772c19566
+oid sha256:438c0b912e69bdcffb4b36884a45fded87e651b5552ce3d28f7d81aadd80e521
 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 sha256:71cdf44715446c06dd4fba004a7a18790345b0ae4b9e2bcb1cb99e4155dbc62d
+oid sha256:29698bcd3e83b59e82799f82508a0c896d6666ff11032f6f0bdfe1ebc23e6468
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3978251516819,
-  "best_model_checkpoint": "mikhail_panzo/ceb_b64_le4_s8000/checkpoint-3000",
-  "epoch": 118.81188118811882,
+  "best_metric": 0.39598318934440613,
+  "best_model_checkpoint": "mikhail_panzo/ceb_b64_le4_s8000/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": 22.033,
       "eval_steps_per_second": 2.815,
       "step": 3000
+    },
+    {
+      "epoch": 120.79207920792079,
+      "grad_norm": 0.8712412118911743,
+      "learning_rate": 8.251666666666668e-05,
+      "loss": 0.3675,
+      "step": 3050
+    },
+    {
+      "epoch": 122.77227722772277,
+      "grad_norm": 4.452208042144775,
+      "learning_rate": 8.168333333333333e-05,
+      "loss": 0.3687,
+      "step": 3100
+    },
+    {
+      "epoch": 124.75247524752476,
+      "grad_norm": 2.735180377960205,
+      "learning_rate": 8.085e-05,
+      "loss": 0.3749,
+      "step": 3150
+    },
+    {
+      "epoch": 126.73267326732673,
+      "grad_norm": 2.1853744983673096,
+      "learning_rate": 8.001666666666667e-05,
+      "loss": 0.3733,
+      "step": 3200
+    },
+    {
+      "epoch": 128.7128712871287,
+      "grad_norm": 3.216191530227661,
+      "learning_rate": 7.918333333333334e-05,
+      "loss": 0.369,
+      "step": 3250
+    },
+    {
+      "epoch": 130.69306930693068,
+      "grad_norm": 1.2702809572219849,
+      "learning_rate": 7.835000000000001e-05,
+      "loss": 0.3673,
+      "step": 3300
+    },
+    {
+      "epoch": 132.67326732673268,
+      "grad_norm": 2.0314784049987793,
+      "learning_rate": 7.751666666666668e-05,
+      "loss": 0.3671,
+      "step": 3350
+    },
+    {
+      "epoch": 134.65346534653466,
+      "grad_norm": 2.0706610679626465,
+      "learning_rate": 7.668333333333335e-05,
+      "loss": 0.3625,
+      "step": 3400
+    },
+    {
+      "epoch": 136.63366336633663,
+      "grad_norm": 1.2799315452575684,
+      "learning_rate": 7.585e-05,
+      "loss": 0.3646,
+      "step": 3450
+    },
+    {
+      "epoch": 138.6138613861386,
+      "grad_norm": 1.2347270250320435,
+      "learning_rate": 7.501666666666667e-05,
+      "loss": 0.365,
+      "step": 3500
+    },
+    {
+      "epoch": 138.6138613861386,
+      "eval_loss": 0.39598318934440613,
+      "eval_runtime": 7.6328,
+      "eval_samples_per_second": 23.582,
+      "eval_steps_per_second": 3.013,
+      "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