Training in progress, step 4100, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:737ac7a6ab5f3b72449f52a2ff1e49910f7e01ca4d76d82317489a184db9c45e
+oid sha256:7dbc1c026e5401052deecba5fb4b823c8759a1342bc564bdd779135c687df62a
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:553a79492198b62da86b2c40e9e863d406fed5b51748de846937ce6ae6e40546
+oid sha256:e1f554142ed5a70cb0c166d54e45a4e8cb5f259d349c1db20c12e03901a577e4
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aba10e0017806276ad36c9962a8d3fa89bfabe64354bedff59216e48a02da40c
+oid sha256:be9cef20fcb5836bc994b3161e3caaa39c6412b1a8cf4e470a339da81589749a
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ab644d38b64875fe4b9db6a2dd788efbee5d9e5dadfabf0dd125e8dae428442
+oid sha256:1ef51ed05818f33aebca9592cb64baf0df006aad2a34a7cf944f4a48b70afce3
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.2116272449493408,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9143897996357016,
+  "best_metric": 1.1990991830825806,
+  "best_model_checkpoint": "./outputs/checkpoint-4100",
+  "epoch": 2.987249544626594,
   "eval_steps": 100,
-  "global_step": 4000,
+  "global_step": 4100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -567,13 +567,27 @@
       "eval_samples_per_second": 21.242,
       "eval_steps_per_second": 2.658,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.1882,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.1990991830825806,
+      "eval_runtime": 133.6346,
+      "eval_samples_per_second": 46.949,
+      "eval_steps_per_second": 5.874,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.3843183295659213e+17,
+  "total_flos": 2.444085005899776e+17,
   "trial_name": null,
   "trial_params": null
 }