Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3335d97bb453ed56dd1661109c866dc67cd3949ab5ea4177066490fb0e6a21f
+oid sha256:2d34f51b826488b7a68e04dd1ad9a2012437084c8afce13bedeacdd8049d3779
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56833e0258e9cffca078067a35e3735ac8fe9bb90a7d7e1861345c8788f9dd56
+oid sha256:3fa3baf33adc9b53e289ad704c1565aa54a2d162d348032a671476fad7d07bb7
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f387f966aecfa76017d726e43b1fa9c3ccecd8af5742b255e2ef66c7eeb645b
+oid sha256:a9ef6c3858e7d84aa5f4da6d612efb7619c9f1a991da1e501295c2a72877c544
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a42424f90eb68391066979b79f87ab18e2fce2a0abc640b46cf1b3f4c12c720a
+oid sha256:593f43b0c83adb27a2db37a6418c2ef12a213bbc2a02f2dc881de6846a69a931
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0614216327667236,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841530054644809,
+  "best_metric": 2.053968906402588,
+  "best_model_checkpoint": "./outputs/checkpoint-4000",
+  "epoch": 2.9143897996357016,
   "eval_steps": 100,
-  "global_step": 3900,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -553,13 +553,27 @@
       "eval_samples_per_second": 30.531,
       "eval_steps_per_second": 3.82,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 2.0234,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 2.053968906402588,
+      "eval_runtime": 205.7574,
+      "eval_samples_per_second": 30.492,
+      "eval_steps_per_second": 3.815,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1105217882630144e+17,
+  "total_flos": 1.1387458434336768e+17,
   "trial_name": null,
   "trial_params": null
 }