Training in progress, step 5500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fb634d25a31951e441158f54ae0db2c273aaeb251169c41f8467cecc648bf56
+oid sha256:226994f9cc097e8b5364a4a2fa612d3987a6cad23e949b085227c213a4385c8c
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cc68ae6c7c867a9ae1fe921933744282f9631aa37051c6ca0a82c81119ae62e
+oid sha256:93f9f2a7999533b71352726a63f81e73d5b01dafaa3480d9c7002eb3b73664dd
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f4e204979057d17355f455adeaa53e6157f86fff91322a4565250d75055d836
+oid sha256:e814e69ac405112ac0e823a174f61a291238359cb5e185b2802f269183d97fb3
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:404094f43137b5b2180fadcde3d1168585bcd71ff155f508576088e11be9d3e1
+oid sha256:59de8a8992a94a360bd3cf9a1be9afda63623261a7fdb31a70c9e1b4bf3d5cb5
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3284249007701874,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-5000",
-  "epoch": 8.37696335078534,
+  "best_metric": 0.3231986165046692,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-5500",
+  "epoch": 9.214659685863875,
   "eval_steps": 500,
-  "global_step": 5000,
+  "global_step": 5500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -787,6 +787,84 @@
       "eval_samples_per_second": 30.901,
       "eval_steps_per_second": 3.866,
       "step": 5000
+    },
+    {
+      "epoch": 8.460732984293193,
+      "grad_norm": 1.1301287412643433,
+      "learning_rate": 4.92e-05,
+      "loss": 0.3542,
+      "step": 5050
+    },
+    {
+      "epoch": 8.544502617801047,
+      "grad_norm": 1.0920559167861938,
+      "learning_rate": 4.836666666666667e-05,
+      "loss": 0.3534,
+      "step": 5100
+    },
+    {
+      "epoch": 8.6282722513089,
+      "grad_norm": 1.1312081813812256,
+      "learning_rate": 4.7533333333333334e-05,
+      "loss": 0.3598,
+      "step": 5150
+    },
+    {
+      "epoch": 8.712041884816754,
+      "grad_norm": 1.5819182395935059,
+      "learning_rate": 4.6700000000000003e-05,
+      "loss": 0.3537,
+      "step": 5200
+    },
+    {
+      "epoch": 8.795811518324607,
+      "grad_norm": 1.0059967041015625,
+      "learning_rate": 4.5866666666666666e-05,
+      "loss": 0.3553,
+      "step": 5250
+    },
+    {
+      "epoch": 8.879581151832461,
+      "grad_norm": 1.1407588720321655,
+      "learning_rate": 4.5033333333333335e-05,
+      "loss": 0.3557,
+      "step": 5300
+    },
+    {
+      "epoch": 8.963350785340314,
+      "grad_norm": 1.296221137046814,
+      "learning_rate": 4.4200000000000004e-05,
+      "loss": 0.3545,
+      "step": 5350
+    },
+    {
+      "epoch": 9.047120418848168,
+      "grad_norm": 1.4306052923202515,
+      "learning_rate": 4.3366666666666666e-05,
+      "loss": 0.3583,
+      "step": 5400
+    },
+    {
+      "epoch": 9.13089005235602,
+      "grad_norm": 1.1458420753479004,
+      "learning_rate": 4.2533333333333335e-05,
+      "loss": 0.3503,
+      "step": 5450
+    },
+    {
+      "epoch": 9.214659685863875,
+      "grad_norm": 0.9508205056190491,
+      "learning_rate": 4.17e-05,
+      "loss": 0.352,
+      "step": 5500
+    },
+    {
+      "epoch": 9.214659685863875,
+      "eval_loss": 0.3231986165046692,
+      "eval_runtime": 273.3343,
+      "eval_samples_per_second": 31.057,
+      "eval_steps_per_second": 3.885,
+      "step": 5500
     }
   ],
   "logging_steps": 50,
@@ -806,7 +884,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.958474220139136e+16,
+  "total_flos": 9.854783789898643e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null