Training in progress, step 3000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b591f8bd066ebc604fbbdad137bf97afb9244c68aa9bb520f2c7090759a4f78
+oid sha256:ca85ee1227af00acd8f2682e5d492914bf2203286c5c539d65f69561c90fe772
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50931ac63787b8d7f0debcc033b1bc61ff3bd92534b9b9e882141449be9287cf
+oid sha256:b5d40362be05624047ee2682319bfc2b5b98f57c50a8e18ab22cf9562c0bd200
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbc76b7bec091f699649ff0404a66c1f836354225ab345408f72d060dc6751c5
+oid sha256:b59f3dcc198907f9a8978b70883ad447ac128644d1bfbfb0001f5dafdb29dfde
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba68065c0ba5863ea7a5974f2a119aa320d107d98a915a0d39227618f3cb2905
+oid sha256:944131a8cfe7c888a8dee4aec652300305a40382e916372dab3fcd26a65bac19
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.41728243231773376,
-  "best_model_checkpoint": "mikhail-panzo/fil_b64_le5_s8000/checkpoint-2500",
-  "epoch": 108.69565217391305,
+  "best_metric": 0.4168698191642761,
+  "best_model_checkpoint": "mikhail-panzo/fil_b64_le5_s8000/checkpoint-3000",
+  "epoch": 130.43478260869566,
   "eval_steps": 500,
-  "global_step": 2500,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -397,6 +397,84 @@
       "eval_samples_per_second": 24.624,
       "eval_steps_per_second": 3.097,
       "step": 2500
+    },
+    {
+      "epoch": 110.8695652173913,
+      "grad_norm": 1.4863485097885132,
+      "learning_rate": 9.088333333333334e-06,
+      "loss": 0.4562,
+      "step": 2550
+    },
+    {
+      "epoch": 113.04347826086956,
+      "grad_norm": 1.1933757066726685,
+      "learning_rate": 9.005000000000001e-06,
+      "loss": 0.4438,
+      "step": 2600
+    },
+    {
+      "epoch": 115.21739130434783,
+      "grad_norm": 2.2257957458496094,
+      "learning_rate": 8.921666666666668e-06,
+      "loss": 0.4511,
+      "step": 2650
+    },
+    {
+      "epoch": 117.3913043478261,
+      "grad_norm": 1.382228970527649,
+      "learning_rate": 8.838333333333335e-06,
+      "loss": 0.4442,
+      "step": 2700
+    },
+    {
+      "epoch": 119.56521739130434,
+      "grad_norm": 1.1577272415161133,
+      "learning_rate": 8.755e-06,
+      "loss": 0.4376,
+      "step": 2750
+    },
+    {
+      "epoch": 121.73913043478261,
+      "grad_norm": 1.16239333152771,
+      "learning_rate": 8.671666666666667e-06,
+      "loss": 0.4378,
+      "step": 2800
+    },
+    {
+      "epoch": 123.91304347826087,
+      "grad_norm": 1.3845179080963135,
+      "learning_rate": 8.588333333333334e-06,
+      "loss": 0.4368,
+      "step": 2850
+    },
+    {
+      "epoch": 126.08695652173913,
+      "grad_norm": 1.4513936042785645,
+      "learning_rate": 8.505e-06,
+      "loss": 0.4362,
+      "step": 2900
+    },
+    {
+      "epoch": 128.2608695652174,
+      "grad_norm": 2.2269773483276367,
+      "learning_rate": 8.421666666666668e-06,
+      "loss": 0.4498,
+      "step": 2950
+    },
+    {
+      "epoch": 130.43478260869566,
+      "grad_norm": 2.3055992126464844,
+      "learning_rate": 8.338333333333335e-06,
+      "loss": 0.4376,
+      "step": 3000
+    },
+    {
+      "epoch": 130.43478260869566,
+      "eval_loss": 0.4168698191642761,
+      "eval_runtime": 6.612,
+      "eval_samples_per_second": 24.047,
+      "eval_steps_per_second": 3.025,
+      "step": 3000
     }
   ],
   "logging_steps": 50,
@@ -416,7 +494,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.710601571250534e+16,
+  "total_flos": 4.453122644505216e+16,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null