Training in progress, step 4500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50723b627f2e210af9d9bfafb7248291159060758af678ccaa8f9b486394ccd6
+oid sha256:17cb28363517fd3837e1f7695aeedfd2415e373d8a816c3e1aab11977228bbb7
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f3d9dd655470cbed1457172994982f38ec142e197173b09f4b533f5abb96af8
+oid sha256:b395277a723ae35d7b37f3cdc7767b97c3a91d943ee0b4afb0b485cc18b6e907
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ff49ebf4516885a27786526fac90152845250ae5c55c1d9becfb25d7b902ac2
+oid sha256:84b43298f2228ef3dc276443417262ccf3c20c277913037cab19e6af2d8cb042
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6441f1f5daac5ee6d48f3a7ecb937462c05426459f6a1b04d27d18ec684da0bb
+oid sha256:2094983ff9f11d6250b785dfe29c8c85823d0967c14aec5d510a4e0433164679
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.41097140312194824,
-  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-4000",
-  "epoch": 173.91304347826087,
+  "best_metric": 0.40913259983062744,
+  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-4500",
+  "epoch": 195.65217391304347,
   "eval_steps": 500,
-  "global_step": 4000,
+  "global_step": 4500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -631,6 +631,84 @@
       "eval_samples_per_second": 12.499,
       "eval_steps_per_second": 1.63,
       "step": 4000
+    },
+    {
+      "epoch": 176.08695652173913,
+      "grad_norm": 0.9379887580871582,
+      "learning_rate": 6.5866666666666666e-06,
+      "loss": 0.426,
+      "step": 4050
+    },
+    {
+      "epoch": 178.2608695652174,
+      "grad_norm": 1.2516026496887207,
+      "learning_rate": 6.5033333333333335e-06,
+      "loss": 0.4244,
+      "step": 4100
+    },
+    {
+      "epoch": 180.43478260869566,
+      "grad_norm": 0.898951530456543,
+      "learning_rate": 6.42e-06,
+      "loss": 0.4212,
+      "step": 4150
+    },
+    {
+      "epoch": 182.6086956521739,
+      "grad_norm": 0.810229480266571,
+      "learning_rate": 6.336666666666667e-06,
+      "loss": 0.4188,
+      "step": 4200
+    },
+    {
+      "epoch": 184.7826086956522,
+      "grad_norm": 0.8796959519386292,
+      "learning_rate": 6.253333333333333e-06,
+      "loss": 0.4196,
+      "step": 4250
+    },
+    {
+      "epoch": 186.95652173913044,
+      "grad_norm": 0.8476919531822205,
+      "learning_rate": 6.17e-06,
+      "loss": 0.4215,
+      "step": 4300
+    },
+    {
+      "epoch": 189.1304347826087,
+      "grad_norm": 0.8933534622192383,
+      "learning_rate": 6.086666666666667e-06,
+      "loss": 0.4209,
+      "step": 4350
+    },
+    {
+      "epoch": 191.30434782608697,
+      "grad_norm": 0.965189516544342,
+      "learning_rate": 6.003333333333334e-06,
+      "loss": 0.4157,
+      "step": 4400
+    },
+    {
+      "epoch": 193.47826086956522,
+      "grad_norm": 0.6911047101020813,
+      "learning_rate": 5.92e-06,
+      "loss": 0.4156,
+      "step": 4450
+    },
+    {
+      "epoch": 195.65217391304347,
+      "grad_norm": 0.98875492811203,
+      "learning_rate": 5.836666666666667e-06,
+      "loss": 0.4185,
+      "step": 4500
+    },
+    {
+      "epoch": 195.65217391304347,
+      "eval_loss": 0.40913259983062744,
+      "eval_runtime": 13.1748,
+      "eval_samples_per_second": 12.22,
+      "eval_steps_per_second": 1.594,
+      "step": 4500
     }
   ],
   "logging_steps": 50,
@@ -650,7 +728,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.847987186698075e+16,
+  "total_flos": 5.453745092390549e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null