Training in progress, step 3600, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a813af9cf45b66ce50f00095a9f8b47d97322ff114e26ca65ae11a03a1e084c7
+oid sha256:648676827f40593439367dacf2ad638522ef14510808c4ba2af6eabf42c8342c
 size 75507072

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:656b60c976bcc62257ceaf6b3ad352af1a1f34984d78e851e2e637ab4a028eea
+oid sha256:98b8c16c0f5f93065962a416a72869294e42de29542f91565e010fff2a0b375f
 size 151034501

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5111e44aa3c7fb3d2679f29ed14de9b378a43fa023e1e3c7c0e2bac399b6ea6f
+oid sha256:63998d98da08278c470436d7f3090e0552bc25f2e2fd93ff495fe8ccda5df6f6
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31f466e63f2d702a9646f61d3cb0499d7a443ca833cfea51694a53eaa24cfd01
+oid sha256:43e3cf8d56a3f083d00cc85544d76ada2f884a1018c8752332d96f2799911117
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.130289077758789,
-  "best_model_checkpoint": "./outputs/checkpoint-3500",
-  "epoch": 2.5500910746812386,
+  "best_metric": 1.120961308479309,
+  "best_model_checkpoint": "./outputs/checkpoint-3600",
+  "epoch": 2.6229508196721314,
   "eval_steps": 100,
-  "global_step": 3500,
+  "global_step": 3600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -497,13 +497,27 @@
       "eval_samples_per_second": 15.015,
       "eval_steps_per_second": 1.879,
       "step": 3500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 3600
+    },
+    {
+      "epoch": 2.62,
+      "eval_loss": 1.120961308479309,
+      "eval_runtime": 438.2292,
+      "eval_samples_per_second": 14.317,
+      "eval_steps_per_second": 1.791,
+      "step": 3600
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.134669408060162e+18,
+  "total_flos": 1.1670242076660173e+18,
   "trial_name": null,
   "trial_params": null
 }