Training in progress, step 4500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ec1f32c6ec559e9e9700642d1c2711f755ed2dea2129a71b0203f0a9f07f89e
+oid sha256:8219fa745d01c552fa70e49ae0b4630b33849d9c0e0aaa82d12bda3be57713b4
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51ca6e02d7d7f88ee019e64afd6f494a9e403b81b590b0a986d5d3a6ae607f9d
+oid sha256:b064d279985bf003683a12da5ac54901dfd34aecfd621706a9db0e14f7d229da
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:764eba3e2ef751a32e1beed18d1b11b8e1d0174936b72cc1bd4d9db171da39e4
+oid sha256:baf4273b0aa00716ec665818b6ea6570dbcadec6f6867e96181a242ab5b090e7
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f3323bc562f08c84a55f57ffe42cad4e899b9bed240a410d83f006b05ef914d
+oid sha256:abd54cd94767f93e5d208ede71f2b5ed0172a8c421b769cc48d09cb32a6db474
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.40923717617988586,
-  "best_model_checkpoint": "mikhail_panzo/fil_b32_le4_s8000/checkpoint-3500",
-  "epoch": 87.91208791208791,
+  "best_metric": 0.4088541567325592,
+  "best_model_checkpoint": "mikhail_panzo/fil_b32_le4_s8000/checkpoint-4500",
+  "epoch": 98.9010989010989,
   "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": 12.097,
       "eval_steps_per_second": 1.578,
       "step": 4000
+    },
+    {
+      "epoch": 89.01098901098901,
+      "grad_norm": 1.8667621612548828,
+      "learning_rate": 6.586666666666666e-05,
+      "loss": 0.3867,
+      "step": 4050
+    },
+    {
+      "epoch": 90.10989010989012,
+      "grad_norm": 1.7244277000427246,
+      "learning_rate": 6.503333333333333e-05,
+      "loss": 0.3802,
+      "step": 4100
+    },
+    {
+      "epoch": 91.20879120879121,
+      "grad_norm": 1.6144648790359497,
+      "learning_rate": 6.42e-05,
+      "loss": 0.3853,
+      "step": 4150
+    },
+    {
+      "epoch": 92.3076923076923,
+      "grad_norm": 1.299013614654541,
+      "learning_rate": 6.336666666666667e-05,
+      "loss": 0.3837,
+      "step": 4200
+    },
+    {
+      "epoch": 93.4065934065934,
+      "grad_norm": 2.6695783138275146,
+      "learning_rate": 6.253333333333333e-05,
+      "loss": 0.3832,
+      "step": 4250
+    },
+    {
+      "epoch": 94.50549450549451,
+      "grad_norm": 1.719193696975708,
+      "learning_rate": 6.170000000000001e-05,
+      "loss": 0.3832,
+      "step": 4300
+    },
+    {
+      "epoch": 95.6043956043956,
+      "grad_norm": 2.300633668899536,
+      "learning_rate": 6.086666666666667e-05,
+      "loss": 0.381,
+      "step": 4350
+    },
+    {
+      "epoch": 96.7032967032967,
+      "grad_norm": 1.0958266258239746,
+      "learning_rate": 6.003333333333334e-05,
+      "loss": 0.3834,
+      "step": 4400
+    },
+    {
+      "epoch": 97.8021978021978,
+      "grad_norm": 1.2469346523284912,
+      "learning_rate": 5.92e-05,
+      "loss": 0.3746,
+      "step": 4450
+    },
+    {
+      "epoch": 98.9010989010989,
+      "grad_norm": 2.032564878463745,
+      "learning_rate": 5.836666666666667e-05,
+      "loss": 0.381,
+      "step": 4500
+    },
+    {
+      "epoch": 98.9010989010989,
+      "eval_loss": 0.4088541567325592,
+      "eval_runtime": 12.5723,
+      "eval_samples_per_second": 12.806,
+      "eval_steps_per_second": 1.67,
+      "step": 4500
     }
   ],
   "logging_steps": 50,
@@ -650,7 +728,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.372276610369547e+16,
+  "total_flos": 2.6691107469196896e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null