Training in progress, epoch 21, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f875ff0bfa8c3f03718200317018a9c1320ef659ed8be49eb8d1545f90dca2b
+oid sha256:759e11911622505bdb5a77d511b901f277ce0bbc06c416da623a53f9c0f0b663
 size 1227009528

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4edde3d63fc51cb76d2b8798e35123bee17f10b37ac6770074f24fbb1849dc32
+oid sha256:c251d2bf032c0284e58db70391eefabbbff2c2fb3bc7e50e9d6ba42d7c0345d5
 size 2454133690

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b0066a3b21610aa70bfcf0b5c4ca5da7f43ab12c9e601ab15813e745474a36d
+oid sha256:4db8bac039de345d3f184975bc589d15c0b93c070db1287ed5147dba9f5ba405
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8c5b80067b711daea816f97793263fb6b6d08534034a3999a4ce7590fa85de8
+oid sha256:e7adf3363afe258d5bb46d306ca21d84331e969bff032715bb85737194722fbc
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 34.54485321044922,
   "best_model_checkpoint": "/kaggle/working/output/checkpoint-20880",
-  "epoch": 20.0,
+  "epoch": 21.0,
   "eval_steps": 500,
-  "global_step": 26100,
+  "global_step": 27405,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1994,6 +1994,105 @@
       "eval_samples_per_second": 26.478,
       "eval_steps_per_second": 3.328,
       "step": 26100
+    },
+    {
+      "epoch": 20.07662835249042,
+      "grad_norm": 4.07724142074585,
+      "learning_rate": 3.745785440613027e-05,
+      "loss": 33.562,
+      "step": 26200
+    },
+    {
+      "epoch": 20.153256704980844,
+      "grad_norm": 4.335379600524902,
+      "learning_rate": 3.7409961685823756e-05,
+      "loss": 33.166,
+      "step": 26300
+    },
+    {
+      "epoch": 20.229885057471265,
+      "grad_norm": 5.472820281982422,
+      "learning_rate": 3.736206896551724e-05,
+      "loss": 33.8918,
+      "step": 26400
+    },
+    {
+      "epoch": 20.306513409961685,
+      "grad_norm": 3.011789321899414,
+      "learning_rate": 3.731417624521073e-05,
+      "loss": 33.395,
+      "step": 26500
+    },
+    {
+      "epoch": 20.38314176245211,
+      "grad_norm": 3.251089096069336,
+      "learning_rate": 3.7266283524904216e-05,
+      "loss": 32.9072,
+      "step": 26600
+    },
+    {
+      "epoch": 20.45977011494253,
+      "grad_norm": 2.7508978843688965,
+      "learning_rate": 3.72183908045977e-05,
+      "loss": 33.92,
+      "step": 26700
+    },
+    {
+      "epoch": 20.53639846743295,
+      "grad_norm": 2.8051536083221436,
+      "learning_rate": 3.717049808429119e-05,
+      "loss": 33.9392,
+      "step": 26800
+    },
+    {
+      "epoch": 20.613026819923373,
+      "grad_norm": 7.377379417419434,
+      "learning_rate": 3.712260536398468e-05,
+      "loss": 33.0382,
+      "step": 26900
+    },
+    {
+      "epoch": 20.689655172413794,
+      "grad_norm": 3.7770464420318604,
+      "learning_rate": 3.7074712643678164e-05,
+      "loss": 32.6836,
+      "step": 27000
+    },
+    {
+      "epoch": 20.766283524904214,
+      "grad_norm": 4.923346996307373,
+      "learning_rate": 3.702681992337165e-05,
+      "loss": 33.2129,
+      "step": 27100
+    },
+    {
+      "epoch": 20.842911877394634,
+      "grad_norm": 4.790703773498535,
+      "learning_rate": 3.697892720306513e-05,
+      "loss": 33.5413,
+      "step": 27200
+    },
+    {
+      "epoch": 20.919540229885058,
+      "grad_norm": 4.592926025390625,
+      "learning_rate": 3.6931034482758624e-05,
+      "loss": 33.2436,
+      "step": 27300
+    },
+    {
+      "epoch": 20.99616858237548,
+      "grad_norm": 3.0529520511627197,
+      "learning_rate": 3.688314176245211e-05,
+      "loss": 33.2415,
+      "step": 27400
+    },
+    {
+      "epoch": 21.0,
+      "eval_loss": 34.59661865234375,
+      "eval_runtime": 49.3345,
+      "eval_samples_per_second": 26.452,
+      "eval_steps_per_second": 3.324,
+      "step": 27405
     }
   ],
   "logging_steps": 100,
@@ -2008,7 +2107,7 @@
         "early_stopping_threshold": 0.0
       },
       "attributes": {
-        "early_stopping_patience_counter": 4
+        "early_stopping_patience_counter": 5
       }
     },
     "TrainerControl": {
@@ -2022,7 +2121,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.814621806094336e+16,
+  "total_flos": 2.955352896399053e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null