Training in progress, epoch 8, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0726fa530c6ac77f8f3c68ffe7f1ba01db45a098cc51f636a7c1ba8b59d3586d
+oid sha256:4c1d72a952aab7cc88a8dbc435765141011109cc7a49170eba6846d78002ebd5
 size 1980860410

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b34dbfea05eb5422e006a144b5f954b538f4359044e7e416f9e704e152a7989b
+oid sha256:9d0d58ebff9eb5d4fe5d50d8af5a5b026bde157c42d02e738198e47ff0208caa
 size 990409330

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:987189384bf09d11dec2af9c70406cd2efd3483f1c3cd71dce5b9fd6c31868a2
+oid sha256:a647527321fbac0898e0e5ee7259accc58bf765276b4863a4b0630dd5450bbe8
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14143acff0ead4485046a1fe7314fe8dc180e8a9f26db4244c2b5510c7a3c384
+oid sha256:42c10f619cac08078ffceaf68eb88edad8b16759e503a95030549ce6ade96b35
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 1.9787917137145996,
   "best_model_checkpoint": "dq158/morbius/checkpoint-790",
-  "epoch": 6.9984192222573505,
+  "epoch": 8.0,
   "eval_steps": 500,
-  "global_step": 2767,
+  "global_step": 3163,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -170,13 +170,38 @@
       "eval_steps_per_second": 0.628,
       "eval_translation_length": 51995,
       "step": 2767
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 5.2069323033477215e-06,
+      "loss": 2.1498,
+      "step": 3000
+    },
+    {
+      "epoch": 8.0,
+      "eval_bleu": 1.0,
+      "eval_brevity_penalty": 1.0,
+      "eval_length_ratio": 1.0,
+      "eval_loss": 1.9796102046966553,
+      "eval_precisions": [
+        1.0,
+        1.0,
+        1.0,
+        1.0
+      ],
+      "eval_reference_length": 51999,
+      "eval_runtime": 559.6356,
+      "eval_samples_per_second": 5.025,
+      "eval_steps_per_second": 0.629,
+      "eval_translation_length": 51999,
+      "step": 3163
     }
   ],
   "logging_steps": 500,
   "max_steps": 7900,
   "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 1.212849070229422e+17,
+  "total_flos": 1.3861132231193395e+17,
   "trial_name": null,
   "trial_params": null
 }