Training in progress, step 5500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e89ff2481874bf269cba7ac84ce957b44c8fa49e2c675e9e6b16008fdb7a55c7
+oid sha256:127785b687ae550620c07c76161ce4fde314a3f786bb3bdecdd030f4cff25442
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b314243ca4b21fc0b329b39dc658d58545fdaa26b0dc513de3ccf6b20a972ec
+oid sha256:f56a73e3c0f71225db11233195f56bf1f4661b8840ebb310a361c5bbb1b1c79b
 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 sha256:51ae9e871201b5140a75803631a3b447a66295ef984c1c9cd1719a8febdf7390
+oid sha256:1f4ebc94f219234dc117648ea36b52869f5ad7c506a78e36de2ec35324b116e9
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.37838345766067505,
-  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_s8000/checkpoint-5000",
-  "epoch": 8.37696335078534,
+  "best_metric": 0.3729405403137207,
+  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_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": 33.136,
       "eval_steps_per_second": 4.145,
       "step": 5000
+    },
+    {
+      "epoch": 8.460732984293193,
+      "grad_norm": 1.079275369644165,
+      "learning_rate": 4.921666666666666e-06,
+      "loss": 0.4143,
+      "step": 5050
+    },
+    {
+      "epoch": 8.544502617801047,
+      "grad_norm": 1.0949628353118896,
+      "learning_rate": 4.838333333333334e-06,
+      "loss": 0.4123,
+      "step": 5100
+    },
+    {
+      "epoch": 8.6282722513089,
+      "grad_norm": 1.5005624294281006,
+      "learning_rate": 4.755e-06,
+      "loss": 0.4218,
+      "step": 5150
+    },
+    {
+      "epoch": 8.712041884816754,
+      "grad_norm": 5.408727169036865,
+      "learning_rate": 4.671666666666667e-06,
+      "loss": 0.4123,
+      "step": 5200
+    },
+    {
+      "epoch": 8.795811518324607,
+      "grad_norm": 1.0426135063171387,
+      "learning_rate": 4.588333333333333e-06,
+      "loss": 0.4159,
+      "step": 5250
+    },
+    {
+      "epoch": 8.879581151832461,
+      "grad_norm": 2.312485456466675,
+      "learning_rate": 4.505e-06,
+      "loss": 0.4149,
+      "step": 5300
+    },
+    {
+      "epoch": 8.963350785340314,
+      "grad_norm": 1.6531749963760376,
+      "learning_rate": 4.421666666666667e-06,
+      "loss": 0.4147,
+      "step": 5350
+    },
+    {
+      "epoch": 9.047120418848168,
+      "grad_norm": 1.2448443174362183,
+      "learning_rate": 4.338333333333334e-06,
+      "loss": 0.4191,
+      "step": 5400
+    },
+    {
+      "epoch": 9.13089005235602,
+      "grad_norm": 1.1506507396697998,
+      "learning_rate": 4.255e-06,
+      "loss": 0.4094,
+      "step": 5450
+    },
+    {
+      "epoch": 9.214659685863875,
+      "grad_norm": 1.2973967790603638,
+      "learning_rate": 4.171666666666667e-06,
+      "loss": 0.412,
+      "step": 5500
+    },
+    {
+      "epoch": 9.214659685863875,
+      "eval_loss": 0.3729405403137207,
+      "eval_runtime": 258.7159,
+      "eval_samples_per_second": 32.812,
+      "eval_steps_per_second": 4.105,
+      "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