Training in progress, step 9000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0f2409f7977c0c11070e7847ab6d58e94975afadfbf2a12bd0548fcd7859a55
+oid sha256:9107c0ff4e2763513ec5bf8850c4b4e538db65052ee12ccd65d557cafb1e34b8
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4133e095e6d740409db22a76e80e3b6a7fa538d8c6ba6fe7edb1a0ede80eaf60
+oid sha256:c1e28ceda6028390e61736b0495cead414f93a42f97c1b979bdc56131341a08c
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:517c8b809d6792d05ae4c9199141efb435941aafdeb5561eaa907592337ad507
+oid sha256:12b7a8f09422f4c6b082af7bb144b5b0532b4a256c4619f625ac9a304c7ea0aa
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5efdf488bd79f4078dcc22ed21cb113945810568eae9749dc0cb12dc690695b7
+oid sha256:4f334224cb0aa3586560a9532b53aefd38c1d84e10d9b30ca047caee043ff116
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.310507208108902,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-8500",
-  "epoch": 14.24083769633508,
+  "best_metric": 0.31039854884147644,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-9000",
+  "epoch": 15.078534031413612,
   "eval_steps": 500,
-  "global_step": 8500,
+  "global_step": 9000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1333,6 +1333,84 @@
       "eval_samples_per_second": 30.365,
       "eval_steps_per_second": 3.799,
       "step": 8500
+    },
+    {
+      "epoch": 14.324607329842932,
+      "grad_norm": 2.537041425704956,
+      "learning_rate": 3.453e-05,
+      "loss": 0.3418,
+      "step": 8550
+    },
+    {
+      "epoch": 14.408376963350785,
+      "grad_norm": 1.3357998132705688,
+      "learning_rate": 3.403e-05,
+      "loss": 0.3408,
+      "step": 8600
+    },
+    {
+      "epoch": 14.492146596858639,
+      "grad_norm": 0.8550173044204712,
+      "learning_rate": 3.353e-05,
+      "loss": 0.3408,
+      "step": 8650
+    },
+    {
+      "epoch": 14.575916230366492,
+      "grad_norm": 1.4455218315124512,
+      "learning_rate": 3.303e-05,
+      "loss": 0.3407,
+      "step": 8700
+    },
+    {
+      "epoch": 14.659685863874346,
+      "grad_norm": 1.0547473430633545,
+      "learning_rate": 3.253e-05,
+      "loss": 0.3382,
+      "step": 8750
+    },
+    {
+      "epoch": 14.743455497382199,
+      "grad_norm": 1.5398694276809692,
+      "learning_rate": 3.2029999999999997e-05,
+      "loss": 0.3402,
+      "step": 8800
+    },
+    {
+      "epoch": 14.827225130890053,
+      "grad_norm": 1.008465051651001,
+      "learning_rate": 3.1530000000000005e-05,
+      "loss": 0.3433,
+      "step": 8850
+    },
+    {
+      "epoch": 14.910994764397905,
+      "grad_norm": 1.8319462537765503,
+      "learning_rate": 3.1030000000000006e-05,
+      "loss": 0.341,
+      "step": 8900
+    },
+    {
+      "epoch": 14.99476439790576,
+      "grad_norm": 1.1432167291641235,
+      "learning_rate": 3.053e-05,
+      "loss": 0.3369,
+      "step": 8950
+    },
+    {
+      "epoch": 15.078534031413612,
+      "grad_norm": 1.098186731338501,
+      "learning_rate": 3.0030000000000002e-05,
+      "loss": 0.3396,
+      "step": 9000
+    },
+    {
+      "epoch": 15.078534031413612,
+      "eval_loss": 0.31039854884147644,
+      "eval_runtime": 280.3967,
+      "eval_samples_per_second": 30.275,
+      "eval_steps_per_second": 3.787,
+      "step": 9000
     }
   ],
   "logging_steps": 50,
@@ -1352,7 +1430,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.522865163316464e+17,
+  "total_flos": 1.6124734632740026e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null