Training in progress, step 2400, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:701f13c64afee1edcc48e3f4fd6ea9864775ca8a64b2f317a6f17366dcd624bc
+oid sha256:03c9a04c93f04730cd34b9de4ff0d00f971c0b828bb532c5cce078890208e48f
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e3e86ed23c756f0c4b1a819e3ad3ad2692bbd02554c2995e6607f5bccc85eff
+oid sha256:1e3b42fb9d8c5f41540580d67d8d8f150641f11dc5a9ea914782fcdc5d2c8f67
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba0d2f71b3765f30a8a92ba02eaf902a1662f276a25395f9d1484f184257488f
+oid sha256:376b77f4eda77baae5bb9c56735abd6b9c105452e0922a23656662b4b444f0bc
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:58c098aa199ff890a6444b804a3fbc89c9468b8e4fb7f53db1b50396e4ed0c2d
+oid sha256:5a78e728336b77d3305afe2734acb1241c76f8f2c33b35460bc07397abc49ba2
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.4250658750534058,
-  "best_model_checkpoint": "./outputs/checkpoint-2300",
-  "epoch": 1.6761384335154828,
+  "best_metric": 1.408768892288208,
+  "best_model_checkpoint": "./outputs/checkpoint-2400",
+  "epoch": 1.7489981785063753,
   "eval_steps": 100,
-  "global_step": 2300,
+  "global_step": 2400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -329,13 +329,27 @@
       "eval_samples_per_second": 46.957,
       "eval_steps_per_second": 5.875,
       "step": 2300
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 2400
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 1.408768892288208,
+      "eval_runtime": 133.5488,
+      "eval_samples_per_second": 46.979,
+      "eval_steps_per_second": 5.878,
+      "step": 2400
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.3710020185554125e+17,
+  "total_flos": 1.4305055049825485e+17,
   "trial_name": null,
   "trial_params": null
 }