Training in progress, step 2500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77cf949a0f97184343b7d38f784b04fdb71aaee58aa4de38e1e5079b54b4ba58
+oid sha256:1b591f8bd066ebc604fbbdad137bf97afb9244c68aa9bb520f2c7090759a4f78
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0048ca8ae538b9b924a62e63ceffd10e5b7e2ad8f482ef7babfb9890a931dd73
+oid sha256:50931ac63787b8d7f0debcc033b1bc61ff3bd92534b9b9e882141449be9287cf
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ab606683dee59e251ef2cbfcb1aebf2f02037a37af81cb49a67b06660659a4f
+oid sha256:dbc76b7bec091f699649ff0404a66c1f836354225ab345408f72d060dc6751c5
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b911007def71d4e0760e095fb38b9695588c94c6dfc0cb1ee70d933b50c5ece7
+oid sha256:ba68065c0ba5863ea7a5974f2a119aa320d107d98a915a0d39227618f3cb2905
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.4230838716030121,
-  "best_model_checkpoint": "mikhail-panzo/fil_b64_le5_s8000/checkpoint-2000",
-  "epoch": 86.95652173913044,
+  "best_metric": 0.41728243231773376,
+  "best_model_checkpoint": "mikhail-panzo/fil_b64_le5_s8000/checkpoint-2500",
+  "epoch": 108.69565217391305,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -319,6 +319,84 @@
       "eval_samples_per_second": 24.57,
       "eval_steps_per_second": 3.091,
       "step": 2000
+    },
+    {
+      "epoch": 89.1304347826087,
+      "grad_norm": 2.8028907775878906,
+      "learning_rate": 9.920000000000002e-06,
+      "loss": 0.4694,
+      "step": 2050
+    },
+    {
+      "epoch": 91.30434782608695,
+      "grad_norm": 0.8982108235359192,
+      "learning_rate": 9.836666666666668e-06,
+      "loss": 0.4495,
+      "step": 2100
+    },
+    {
+      "epoch": 93.47826086956522,
+      "grad_norm": 2.500758409500122,
+      "learning_rate": 9.753333333333335e-06,
+      "loss": 0.4456,
+      "step": 2150
+    },
+    {
+      "epoch": 95.65217391304348,
+      "grad_norm": 1.7864420413970947,
+      "learning_rate": 9.67e-06,
+      "loss": 0.4816,
+      "step": 2200
+    },
+    {
+      "epoch": 97.82608695652173,
+      "grad_norm": 1.3289756774902344,
+      "learning_rate": 9.586666666666667e-06,
+      "loss": 0.4512,
+      "step": 2250
+    },
+    {
+      "epoch": 100.0,
+      "grad_norm": 1.5230562686920166,
+      "learning_rate": 9.503333333333334e-06,
+      "loss": 0.4564,
+      "step": 2300
+    },
+    {
+      "epoch": 102.17391304347827,
+      "grad_norm": 2.709244728088379,
+      "learning_rate": 9.42e-06,
+      "loss": 0.4543,
+      "step": 2350
+    },
+    {
+      "epoch": 104.34782608695652,
+      "grad_norm": 2.3624541759490967,
+      "learning_rate": 9.336666666666666e-06,
+      "loss": 0.4505,
+      "step": 2400
+    },
+    {
+      "epoch": 106.52173913043478,
+      "grad_norm": 1.4635056257247925,
+      "learning_rate": 9.253333333333333e-06,
+      "loss": 0.4558,
+      "step": 2450
+    },
+    {
+      "epoch": 108.69565217391305,
+      "grad_norm": 1.9029183387756348,
+      "learning_rate": 9.17e-06,
+      "loss": 0.443,
+      "step": 2500
+    },
+    {
+      "epoch": 108.69565217391305,
+      "eval_loss": 0.41728243231773376,
+      "eval_runtime": 6.4572,
+      "eval_samples_per_second": 24.624,
+      "eval_steps_per_second": 3.097,
+      "step": 2500
     }
   ],
   "logging_steps": 50,
@@ -338,7 +416,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.9689323880306464e+16,
+  "total_flos": 3.710601571250534e+16,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null