Training in progress, step 6000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:226994f9cc097e8b5364a4a2fa612d3987a6cad23e949b085227c213a4385c8c
+oid sha256:8d2b3399969fb74960ee01ffd897aef3b9bf5e7209f6f0ea877a12153f361408
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93f9f2a7999533b71352726a63f81e73d5b01dafaa3480d9c7002eb3b73664dd
+oid sha256:7746ae2b28b16ee5e4dc0d43c7ed500663b9d82298a7c15d1500a39d65d9d274
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e814e69ac405112ac0e823a174f61a291238359cb5e185b2802f269183d97fb3
+oid sha256:8b78a35397517539ceb5abaec4c078472043c61c90e9313f43ee762be5908798
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59de8a8992a94a360bd3cf9a1be9afda63623261a7fdb31a70c9e1b4bf3d5cb5
+oid sha256:5cc3a4f205edd4add2b834d8b0c7057cd9c53044c257143116c01897f3cb6067
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3231986165046692,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-5500",
-  "epoch": 9.214659685863875,
+  "best_metric": 0.3189197778701782,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-6000",
+  "epoch": 10.052356020942408,
   "eval_steps": 500,
-  "global_step": 5500,
+  "global_step": 6000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -865,6 +865,84 @@
       "eval_samples_per_second": 31.057,
       "eval_steps_per_second": 3.885,
       "step": 5500
+    },
+    {
+      "epoch": 9.298429319371728,
+      "grad_norm": 2.127049446105957,
+      "learning_rate": 4.086666666666667e-05,
+      "loss": 0.3547,
+      "step": 5550
+    },
+    {
+      "epoch": 9.38219895287958,
+      "grad_norm": 1.5531483888626099,
+      "learning_rate": 4.0033333333333335e-05,
+      "loss": 0.3576,
+      "step": 5600
+    },
+    {
+      "epoch": 9.465968586387435,
+      "grad_norm": 1.5712753534317017,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.3492,
+      "step": 5650
+    },
+    {
+      "epoch": 9.549738219895287,
+      "grad_norm": 1.1668370962142944,
+      "learning_rate": 3.8366666666666666e-05,
+      "loss": 0.3529,
+      "step": 5700
+    },
+    {
+      "epoch": 9.633507853403142,
+      "grad_norm": 1.1639858484268188,
+      "learning_rate": 3.7533333333333335e-05,
+      "loss": 0.3526,
+      "step": 5750
+    },
+    {
+      "epoch": 9.717277486910994,
+      "grad_norm": 1.4188268184661865,
+      "learning_rate": 3.6700000000000004e-05,
+      "loss": 0.3527,
+      "step": 5800
+    },
+    {
+      "epoch": 9.801047120418849,
+      "grad_norm": 1.8606890439987183,
+      "learning_rate": 3.586666666666667e-05,
+      "loss": 0.3517,
+      "step": 5850
+    },
+    {
+      "epoch": 9.884816753926701,
+      "grad_norm": 1.2933155298233032,
+      "learning_rate": 3.5033333333333336e-05,
+      "loss": 0.3463,
+      "step": 5900
+    },
+    {
+      "epoch": 9.968586387434556,
+      "grad_norm": 1.4728389978408813,
+      "learning_rate": 3.4200000000000005e-05,
+      "loss": 0.3505,
+      "step": 5950
+    },
+    {
+      "epoch": 10.052356020942408,
+      "grad_norm": 1.6055335998535156,
+      "learning_rate": 3.336666666666667e-05,
+      "loss": 0.3471,
+      "step": 6000
+    },
+    {
+      "epoch": 10.052356020942408,
+      "eval_loss": 0.3189197778701782,
+      "eval_runtime": 273.3525,
+      "eval_samples_per_second": 31.055,
+      "eval_steps_per_second": 3.885,
+      "step": 6000
     }
   ],
   "logging_steps": 50,
@@ -884,7 +962,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 9.854783789898643e+16,
+  "total_flos": 1.0748974547355264e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null